voice-router-dev 0.8.2 → 0.8.4

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/dist/index.js CHANGED
@@ -182,7 +182,6 @@ __export(src_exports, {
182
182
  createOpenAIWhisperAdapter: () => createOpenAIWhisperAdapter,
183
183
  createSonioxAdapter: () => createSonioxAdapter,
184
184
  createSpeechmaticsAdapter: () => createSpeechmaticsAdapter,
185
- createTemporaryToken: () => createTemporaryToken,
186
185
  createTranscript: () => createTranscript,
187
186
  createTranscription: () => createTranscription,
188
187
  createVoiceRouter: () => createVoiceRouter,
@@ -234,6 +233,9 @@ __export(src_exports, {
234
233
  transcriptionsGet: () => transcriptionsGet,
235
234
  transcriptionsList: () => transcriptionsList,
236
235
  transcriptionsListFiles: () => transcriptionsListFiles,
236
+ webHooksCreate: () => webHooksCreate,
237
+ webHooksDelete: () => webHooksDelete,
238
+ webHooksList: () => webHooksList,
237
239
  zodToFieldConfigs: () => zodToFieldConfigs
238
240
  });
239
241
  module.exports = __toCommonJS(src_exports);
@@ -2479,13 +2481,6 @@ var TranslationLanguageCodeEnum = {
2479
2481
  zh: "zh"
2480
2482
  };
2481
2483
 
2482
- // src/generated/assemblyai/schema/speechModel.ts
2483
- var SpeechModel = {
2484
- best: "best",
2485
- "slam-1": "slam-1",
2486
- universal: "universal"
2487
- };
2488
-
2489
2484
  // src/generated/assemblyai/schema/transcriptLanguageCode.ts
2490
2485
  var TranscriptLanguageCode = {
2491
2486
  en: "en",
@@ -2801,7 +2796,10 @@ var AssemblyAIEncoding = {
2801
2796
  /** μ-law (telephony) */
2802
2797
  pcmMulaw: "pcm_mulaw"
2803
2798
  };
2804
- var AssemblyAITranscriptionModel = SpeechModel;
2799
+ var AssemblyAITranscriptionModel = {
2800
+ "universal-3-pro": "universal-3-pro",
2801
+ "universal-2": "universal-2"
2802
+ };
2805
2803
  var AssemblyAILanguage = TranscriptLanguageCode;
2806
2804
  var AssemblyAISpeechModel = {
2807
2805
  /** Optimized for English */
@@ -2922,6 +2920,12 @@ var ERROR_CODES = {
2922
2920
  CONNECTION_TIMEOUT: "CONNECTION_TIMEOUT",
2923
2921
  /** Invalid input provided to API */
2924
2922
  INVALID_INPUT: "INVALID_INPUT",
2923
+ /** Authentication failed (invalid or missing API key) */
2924
+ AUTHENTICATION_ERROR: "AUTHENTICATION_ERROR",
2925
+ /** Rate limit exceeded */
2926
+ RATE_LIMIT: "RATE_LIMIT",
2927
+ /** Provider server error (5xx) */
2928
+ SERVER_ERROR: "SERVER_ERROR",
2925
2929
  /** Requested operation not supported by provider */
2926
2930
  NOT_SUPPORTED: "NOT_SUPPORTED",
2927
2931
  /** No transcription results available */
@@ -2936,6 +2940,9 @@ var ERROR_MESSAGES = {
2936
2940
  TRANSCRIPTION_ERROR: "Transcription processing failed",
2937
2941
  CONNECTION_TIMEOUT: "Connection attempt timed out",
2938
2942
  INVALID_INPUT: "Invalid input provided",
2943
+ AUTHENTICATION_ERROR: "Authentication failed (invalid or missing API key)",
2944
+ RATE_LIMIT: "Rate limit exceeded",
2945
+ SERVER_ERROR: "Provider server error",
2939
2946
  NOT_SUPPORTED: "Operation not supported by this provider",
2940
2947
  NO_RESULTS: "No transcription results available",
2941
2948
  UNKNOWN_ERROR: "An unknown error occurred"
@@ -2947,6 +2954,36 @@ function createError(code, customMessage, details) {
2947
2954
  details
2948
2955
  };
2949
2956
  }
2957
+ function httpStatusToErrorCode(status) {
2958
+ switch (status) {
2959
+ case 400:
2960
+ case 404:
2961
+ case 422:
2962
+ return ERROR_CODES.INVALID_INPUT;
2963
+ case 401:
2964
+ case 403:
2965
+ return ERROR_CODES.AUTHENTICATION_ERROR;
2966
+ case 408:
2967
+ return ERROR_CODES.CONNECTION_TIMEOUT;
2968
+ case 429:
2969
+ return ERROR_CODES.RATE_LIMIT;
2970
+ default:
2971
+ if (status >= 500) return ERROR_CODES.SERVER_ERROR;
2972
+ return ERROR_CODES.UNKNOWN_ERROR;
2973
+ }
2974
+ }
2975
+ function extractProviderMessage(data) {
2976
+ if (!data || typeof data !== "object") {
2977
+ return typeof data === "string" ? data : void 0;
2978
+ }
2979
+ const d = data;
2980
+ if (d.error && typeof d.error === "object" && d.error.message) return String(d.error.message);
2981
+ if (typeof d.error === "string") return d.error;
2982
+ if (d.detail && typeof d.detail === "object" && d.detail.message) return String(d.detail.message);
2983
+ if (typeof d.message === "string") return d.message;
2984
+ if (typeof d.err_msg === "string") return d.err_msg;
2985
+ return void 0;
2986
+ }
2950
2987
 
2951
2988
  // src/adapters/base-adapter.ts
2952
2989
  var BaseAdapter = class {
@@ -2965,12 +3002,15 @@ var BaseAdapter = class {
2965
3002
  const httpStatus = statusCode || err.statusCode || err.response?.status;
2966
3003
  const httpStatusText = err.response?.statusText;
2967
3004
  const responseData = err.response?.data;
3005
+ const errorCode = code || (httpStatus ? httpStatusToErrorCode(httpStatus) : void 0) || ERROR_CODES.UNKNOWN_ERROR;
3006
+ const providerMessage = extractProviderMessage(responseData);
3007
+ const message = providerMessage || err.message || "An unknown error occurred";
2968
3008
  return {
2969
3009
  success: false,
2970
3010
  provider: this.name,
2971
3011
  error: {
2972
- code: code || err.code || ERROR_CODES.UNKNOWN_ERROR,
2973
- message: err.message || "An unknown error occurred",
3012
+ code: errorCode,
3013
+ message,
2974
3014
  statusCode: httpStatus,
2975
3015
  details: {
2976
3016
  // Include full error object
@@ -5384,19 +5424,20 @@ var schema_exports2 = {};
5384
5424
  __export(schema_exports2, {
5385
5425
  AudioIntelligenceModelStatus: () => AudioIntelligenceModelStatus,
5386
5426
  EntityType: () => EntityType,
5387
- LemurModel: () => LemurModel,
5388
5427
  PiiPolicy: () => PiiPolicy,
5389
5428
  RedactPiiAudioQuality: () => RedactPiiAudioQuality,
5390
5429
  RedactedAudioStatus: () => RedactedAudioStatus,
5391
5430
  Sentiment: () => Sentiment,
5392
- SpeechModel: () => SpeechModel,
5431
+ SpeakerIdentificationRequestBodySpeakerIdentificationSpeakerType: () => SpeakerIdentificationRequestBodySpeakerIdentificationSpeakerType,
5393
5432
  SubstitutionPolicy: () => SubstitutionPolicy,
5394
5433
  SubtitleFormat: () => SubtitleFormat,
5395
5434
  SummaryModel: () => SummaryModel,
5396
5435
  SummaryType: () => SummaryType,
5397
- TranscriptBoostParam: () => TranscriptBoostParam,
5398
5436
  TranscriptLanguageCode: () => TranscriptLanguageCode,
5399
- TranscriptReadyStatus: () => TranscriptReadyStatus,
5437
+ TranscriptOptionalParamsRedactPiiAudioOptionsOverrideAudioRedactionMethod: () => TranscriptOptionalParamsRedactPiiAudioOptionsOverrideAudioRedactionMethod,
5438
+ TranscriptOptionalParamsRemoveAudioTags: () => TranscriptOptionalParamsRemoveAudioTags,
5439
+ TranscriptRedactPiiAudioOptionsOverrideAudioRedactionMethod: () => TranscriptRedactPiiAudioOptionsOverrideAudioRedactionMethod,
5440
+ TranscriptRemoveAudioTags: () => TranscriptRemoveAudioTags,
5400
5441
  TranscriptStatus: () => TranscriptStatus
5401
5442
  });
5402
5443
 
@@ -5454,13 +5495,6 @@ var EntityType = {
5454
5495
  zodiac_sign: "zodiac_sign"
5455
5496
  };
5456
5497
 
5457
- // src/generated/assemblyai/schema/lemurModel.ts
5458
- var LemurModel = {
5459
- "anthropic/claude-3-5-sonnet": "anthropic/claude-3-5-sonnet",
5460
- "anthropic/claude-3-opus": "anthropic/claude-3-opus",
5461
- "anthropic/claude-3-haiku": "anthropic/claude-3-haiku"
5462
- };
5463
-
5464
5498
  // src/generated/assemblyai/schema/piiPolicy.ts
5465
5499
  var PiiPolicy = {
5466
5500
  account_number: "account_number",
@@ -5527,6 +5561,12 @@ var Sentiment = {
5527
5561
  NEGATIVE: "NEGATIVE"
5528
5562
  };
5529
5563
 
5564
+ // src/generated/assemblyai/schema/speakerIdentificationRequestBodySpeakerIdentificationSpeakerType.ts
5565
+ var SpeakerIdentificationRequestBodySpeakerIdentificationSpeakerType = {
5566
+ role: "role",
5567
+ name: "name"
5568
+ };
5569
+
5530
5570
  // src/generated/assemblyai/schema/substitutionPolicy.ts
5531
5571
  var SubstitutionPolicy = {
5532
5572
  entity_name: "entity_name",
@@ -5555,17 +5595,24 @@ var SummaryType = {
5555
5595
  paragraph: "paragraph"
5556
5596
  };
5557
5597
 
5558
- // src/generated/assemblyai/schema/transcriptBoostParam.ts
5559
- var TranscriptBoostParam = {
5560
- low: "low",
5561
- default: "default",
5562
- high: "high"
5598
+ // src/generated/assemblyai/schema/transcriptOptionalParamsRedactPiiAudioOptionsOverrideAudioRedactionMethod.ts
5599
+ var TranscriptOptionalParamsRedactPiiAudioOptionsOverrideAudioRedactionMethod = {
5600
+ silence: "silence"
5563
5601
  };
5564
5602
 
5565
- // src/generated/assemblyai/schema/transcriptReadyStatus.ts
5566
- var TranscriptReadyStatus = {
5567
- completed: "completed",
5568
- error: "error"
5603
+ // src/generated/assemblyai/schema/transcriptOptionalParamsRemoveAudioTags.ts
5604
+ var TranscriptOptionalParamsRemoveAudioTags = {
5605
+ all: "all"
5606
+ };
5607
+
5608
+ // src/generated/assemblyai/schema/transcriptRedactPiiAudioOptionsOverrideAudioRedactionMethod.ts
5609
+ var TranscriptRedactPiiAudioOptionsOverrideAudioRedactionMethod = {
5610
+ silence: "silence"
5611
+ };
5612
+
5613
+ // src/generated/assemblyai/schema/transcriptRemoveAudioTags.ts
5614
+ var TranscriptRemoveAudioTags = {
5615
+ all: "all"
5569
5616
  };
5570
5617
 
5571
5618
  // src/generated/assemblyai/api/assemblyAIAPI.ts
@@ -5584,9 +5631,6 @@ var getTranscript = (transcriptId, options) => {
5584
5631
  var deleteTranscript = (transcriptId, options) => {
5585
5632
  return import_axios2.default.delete(`/v2/transcript/${transcriptId}`, options);
5586
5633
  };
5587
- var createTemporaryToken = (createRealtimeTemporaryTokenParams, options) => {
5588
- return import_axios2.default.post(`/v2/realtime/token`, createRealtimeTemporaryTokenParams, options);
5589
- };
5590
5634
 
5591
5635
  // src/adapters/assemblyai-adapter.ts
5592
5636
  var AssemblyAIAdapter = class extends BaseAdapter {
@@ -5611,14 +5655,61 @@ var AssemblyAIAdapter = class extends BaseAdapter {
5611
5655
  this.wsBaseUrl = "wss://streaming.assemblyai.com/v3/ws";
5612
5656
  }
5613
5657
  // v3 Universal Streaming endpoint
5658
+ /**
5659
+ * Get regional hosts for AssemblyAI
5660
+ *
5661
+ * @param region - Regional endpoint identifier
5662
+ * @returns Object with api and streaming hosts
5663
+ */
5664
+ getRegionalHosts(region) {
5665
+ if (region === "eu") {
5666
+ return { api: "api.eu.assemblyai.com", streaming: "streaming.eu.assemblyai.com" };
5667
+ }
5668
+ return { api: "api.assemblyai.com", streaming: "streaming.assemblyai.com" };
5669
+ }
5614
5670
  initialize(config) {
5615
5671
  super.initialize(config);
5616
- if (config.wsBaseUrl) {
5617
- this.wsBaseUrl = config.wsBaseUrl;
5618
- } else if (config.baseUrl) {
5619
- this.wsBaseUrl = `${this.deriveWsUrl(config.baseUrl)}/v3/ws`;
5672
+ const hosts = this.getRegionalHosts(config.region);
5673
+ this.baseUrl = config.baseUrl || `https://${hosts.api}`;
5674
+ this.wsBaseUrl = config.wsBaseUrl || (config.baseUrl ? `${this.deriveWsUrl(config.baseUrl)}/v3/ws` : `wss://${hosts.streaming}/v3/ws`);
5675
+ }
5676
+ /**
5677
+ * Change the regional endpoint dynamically
5678
+ *
5679
+ * Useful for switching between US and EU endpoints without reinitializing.
5680
+ * Affects both REST API and WebSocket streaming endpoints.
5681
+ *
5682
+ * @param region - New regional endpoint to use (`us` or `eu`)
5683
+ *
5684
+ * @example Switch to EU region
5685
+ * ```typescript
5686
+ * import { AssemblyAIRegion } from 'voice-router-dev/constants'
5687
+ *
5688
+ * adapter.setRegion(AssemblyAIRegion.eu)
5689
+ * await adapter.transcribe(audio) // Uses EU endpoint
5690
+ * ```
5691
+ */
5692
+ setRegion(region) {
5693
+ this.validateConfig();
5694
+ if (!this.config.baseUrl) {
5695
+ const hosts = this.getRegionalHosts(region);
5696
+ this.baseUrl = `https://${hosts.api}`;
5697
+ if (!this.config.wsBaseUrl) {
5698
+ this.wsBaseUrl = `wss://${hosts.streaming}/v3/ws`;
5699
+ }
5620
5700
  }
5621
5701
  }
5702
+ /**
5703
+ * Get the current regional endpoints being used
5704
+ *
5705
+ * @returns Object with current API and WebSocket URLs
5706
+ */
5707
+ getRegion() {
5708
+ return {
5709
+ api: this.baseUrl,
5710
+ websocket: this.wsBaseUrl
5711
+ };
5712
+ }
5622
5713
  /**
5623
5714
  * Get axios config for generated API client functions
5624
5715
  * Configures headers and base URL using authorization header
@@ -5878,13 +5969,15 @@ var AssemblyAIAdapter = class extends BaseAdapter {
5878
5969
  const request = {
5879
5970
  ...options?.assemblyai,
5880
5971
  audio_url: audioUrl,
5972
+ // speech_models is required — default to universal-3-pro
5973
+ speech_models: options?.assemblyai?.speech_models ?? ["universal-3-pro"],
5881
5974
  // Enable punctuation and formatting by default
5882
5975
  punctuate: options?.assemblyai?.punctuate ?? true,
5883
5976
  format_text: options?.assemblyai?.format_text ?? true
5884
5977
  };
5885
5978
  if (options) {
5886
5979
  if (options.model) {
5887
- request.speech_model = options.model;
5980
+ request.speech_models = [options.model];
5888
5981
  }
5889
5982
  if (options.language) {
5890
5983
  const languageCode = options.language.includes("_") ? options.language : `${options.language}_us`;
@@ -5900,8 +5993,7 @@ var AssemblyAIAdapter = class extends BaseAdapter {
5900
5993
  }
5901
5994
  }
5902
5995
  if (options.customVocabulary && options.customVocabulary.length > 0) {
5903
- request.word_boost = options.customVocabulary;
5904
- request.boost_param = request.boost_param ?? "high";
5996
+ request.keyterms_prompt = options.customVocabulary;
5905
5997
  }
5906
5998
  if (options.summarization) {
5907
5999
  request.summarization = true;
@@ -7676,6 +7768,18 @@ var transcriptionsListFiles = (id, params, options) => {
7676
7768
  params: { ...params, ...options?.params }
7677
7769
  });
7678
7770
  };
7771
+ var webHooksList = (params, options) => {
7772
+ return import_axios4.default.get(`/webhooks`, {
7773
+ ...options,
7774
+ params: { ...params, ...options?.params }
7775
+ });
7776
+ };
7777
+ var webHooksCreate = (webHook, options) => {
7778
+ return import_axios4.default.post(`/webhooks`, webHook, options);
7779
+ };
7780
+ var webHooksDelete = (id, options) => {
7781
+ return import_axios4.default.delete(`/webhooks/${id}`, options);
7782
+ };
7679
7783
 
7680
7784
  // src/adapters/azure-stt-adapter.ts
7681
7785
  var AzureSTTAdapter = class extends BaseAdapter {
@@ -7748,19 +7852,8 @@ var AzureSTTAdapter = class extends BaseAdapter {
7748
7852
  this.getAxiosConfig()
7749
7853
  );
7750
7854
  const transcription = response.data;
7751
- return {
7752
- success: true,
7753
- provider: this.name,
7754
- data: {
7755
- id: transcription.self?.split("/").pop() || "",
7756
- text: "",
7757
- // Will be populated after polling
7758
- status: this.normalizeStatus(transcription.status),
7759
- language: transcription.locale,
7760
- createdAt: transcription.createdDateTime
7761
- },
7762
- raw: transcription
7763
- };
7855
+ const transcriptId = transcription.self?.split("/").pop() || "";
7856
+ return await this.pollForCompletion(transcriptId);
7764
7857
  } catch (error) {
7765
7858
  return this.createErrorResponse(error);
7766
7859
  }
@@ -7918,6 +8011,50 @@ var AzureSTTAdapter = class extends BaseAdapter {
7918
8011
  };
7919
8012
  }
7920
8013
  }
8014
+ /**
8015
+ * Register a subscription-wide webhook for transcription events
8016
+ *
8017
+ * Azure webhooks are subscription-wide (not per-transcription).
8018
+ * Call this once during setup to receive callbacks for all transcription events.
8019
+ * The webhook URL will receive POST requests for transcription lifecycle events.
8020
+ *
8021
+ * @param url - The webhook URL to receive events
8022
+ * @param options - Optional: event filters and display name
8023
+ * @returns Created webhook object
8024
+ */
8025
+ async registerWebhook(url, options) {
8026
+ this.validateConfig();
8027
+ const webhook = {
8028
+ webUrl: url,
8029
+ displayName: options?.displayName || "SDK Webhook",
8030
+ events: options?.events || {
8031
+ transcriptionCreation: true,
8032
+ transcriptionProcessing: true,
8033
+ transcriptionCompletion: true
8034
+ }
8035
+ };
8036
+ const response = await webHooksCreate(webhook, this.getAxiosConfig());
8037
+ return response.data;
8038
+ }
8039
+ /**
8040
+ * Unregister a subscription-wide webhook by ID
8041
+ *
8042
+ * @param webhookId - The webhook ID to delete
8043
+ */
8044
+ async unregisterWebhook(webhookId) {
8045
+ this.validateConfig();
8046
+ await webHooksDelete(webhookId, this.getAxiosConfig());
8047
+ }
8048
+ /**
8049
+ * List all registered webhooks for the subscription
8050
+ *
8051
+ * @returns Array of registered webhooks
8052
+ */
8053
+ async listWebhooks() {
8054
+ this.validateConfig();
8055
+ const response = await webHooksList(void 0, this.getAxiosConfig());
8056
+ return [...response.data.values || []];
8057
+ }
7921
8058
  /**
7922
8059
  * Map unified status to Azure status format using generated enum
7923
8060
  */
@@ -8674,6 +8811,20 @@ function createOpenAIWhisperAdapter(config) {
8674
8811
  // src/adapters/speechmatics-adapter.ts
8675
8812
  var import_axios8 = __toESM(require("axios"));
8676
8813
 
8814
+ // src/generated/speechmatics/schema/notificationConfigContentsItem.ts
8815
+ var NotificationConfigContentsItem = {
8816
+ jobinfo: "jobinfo",
8817
+ transcript: "transcript",
8818
+ "transcriptjson-v2": "transcript.json-v2",
8819
+ transcripttxt: "transcript.txt",
8820
+ transcriptsrt: "transcript.srt",
8821
+ alignment: "alignment",
8822
+ alignmentword_start_and_end: "alignment.word_start_and_end",
8823
+ alignmentone_per_line: "alignment.one_per_line",
8824
+ data: "data",
8825
+ text: "text"
8826
+ };
8827
+
8677
8828
  // src/generated/speechmatics/schema/transcriptionConfigDiarization.ts
8678
8829
  var TranscriptionConfigDiarization = {
8679
8830
  none: "none",
@@ -8830,6 +8981,14 @@ var SpeechmaticsAdapter = class extends BaseAdapter {
8830
8981
  content: word
8831
8982
  }));
8832
8983
  }
8984
+ if (options?.webhookUrl) {
8985
+ jobConfig.notification_config = [
8986
+ {
8987
+ url: options.webhookUrl,
8988
+ contents: [NotificationConfigContentsItem.transcript]
8989
+ }
8990
+ ];
8991
+ }
8833
8992
  let requestBody;
8834
8993
  let headers = {};
8835
8994
  if (audio.type === "url") {
@@ -8855,16 +9014,20 @@ var SpeechmaticsAdapter = class extends BaseAdapter {
8855
9014
  };
8856
9015
  }
8857
9016
  const response = await this.client.post("/jobs", requestBody, { headers });
8858
- return {
8859
- success: true,
8860
- provider: this.name,
8861
- data: {
8862
- id: response.data.id,
8863
- text: "",
8864
- status: "queued"
8865
- },
8866
- raw: response.data
8867
- };
9017
+ const jobId = response.data.id;
9018
+ if (options?.webhookUrl) {
9019
+ return {
9020
+ success: true,
9021
+ provider: this.name,
9022
+ data: {
9023
+ id: jobId,
9024
+ text: "",
9025
+ status: "queued"
9026
+ },
9027
+ raw: response.data
9028
+ };
9029
+ }
9030
+ return await this.pollForCompletion(jobId);
8868
9031
  } catch (error) {
8869
9032
  return this.createErrorResponse(error);
8870
9033
  }
@@ -11610,15 +11773,10 @@ var deepgramStreamingOnlyParams = import_zod2.z.object({
11610
11773
  // src/generated/assemblyai/api/assemblyAIAPI.zod.ts
11611
11774
  var assemblyAIAPI_zod_exports = {};
11612
11775
  __export(assemblyAIAPI_zod_exports, {
11613
- createTemporaryTokenBody: () => createTemporaryTokenBody,
11614
- createTemporaryTokenBodyExpiresInMin: () => createTemporaryTokenBodyExpiresInMin,
11615
- createTemporaryTokenResponse: () => createTemporaryTokenResponse,
11616
11776
  createTranscriptBody: () => createTranscriptBody,
11617
11777
  createTranscriptBodyAutoChaptersDefault: () => createTranscriptBodyAutoChaptersDefault,
11618
11778
  createTranscriptBodyAutoHighlightsDefault: () => createTranscriptBodyAutoHighlightsDefault,
11619
11779
  createTranscriptBodyContentSafetyConfidenceDefault: () => createTranscriptBodyContentSafetyConfidenceDefault,
11620
- createTranscriptBodyContentSafetyConfidenceMax: () => createTranscriptBodyContentSafetyConfidenceMax,
11621
- createTranscriptBodyContentSafetyConfidenceMin: () => createTranscriptBodyContentSafetyConfidenceMin,
11622
11780
  createTranscriptBodyContentSafetyDefault: () => createTranscriptBodyContentSafetyDefault,
11623
11781
  createTranscriptBodyCustomTopicsDefault: () => createTranscriptBodyCustomTopicsDefault,
11624
11782
  createTranscriptBodyDisfluenciesDefault: () => createTranscriptBodyDisfluenciesDefault,
@@ -11626,198 +11784,58 @@ __export(assemblyAIAPI_zod_exports, {
11626
11784
  createTranscriptBodyFilterProfanityDefault: () => createTranscriptBodyFilterProfanityDefault,
11627
11785
  createTranscriptBodyFormatTextDefault: () => createTranscriptBodyFormatTextDefault,
11628
11786
  createTranscriptBodyIabCategoriesDefault: () => createTranscriptBodyIabCategoriesDefault,
11629
- createTranscriptBodyLanguageCodeDefault: () => createTranscriptBodyLanguageCodeDefault,
11630
- createTranscriptBodyLanguageConfidenceThresholdDefault: () => createTranscriptBodyLanguageConfidenceThresholdDefault,
11631
- createTranscriptBodyLanguageConfidenceThresholdMax: () => createTranscriptBodyLanguageConfidenceThresholdMax,
11632
- createTranscriptBodyLanguageConfidenceThresholdMin: () => createTranscriptBodyLanguageConfidenceThresholdMin,
11633
11787
  createTranscriptBodyLanguageDetectionDefault: () => createTranscriptBodyLanguageDetectionDefault,
11788
+ createTranscriptBodyLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault: () => createTranscriptBodyLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault,
11789
+ createTranscriptBodyLanguageDetectionOptionsCodeSwitchingDefault: () => createTranscriptBodyLanguageDetectionOptionsCodeSwitchingDefault,
11790
+ createTranscriptBodyLanguageDetectionOptionsFallbackLanguageDefault: () => createTranscriptBodyLanguageDetectionOptionsFallbackLanguageDefault,
11634
11791
  createTranscriptBodyMultichannelDefault: () => createTranscriptBodyMultichannelDefault,
11635
11792
  createTranscriptBodyPunctuateDefault: () => createTranscriptBodyPunctuateDefault,
11636
11793
  createTranscriptBodyRedactPiiAudioDefault: () => createTranscriptBodyRedactPiiAudioDefault,
11794
+ createTranscriptBodyRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault: () => createTranscriptBodyRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault,
11637
11795
  createTranscriptBodyRedactPiiDefault: () => createTranscriptBodyRedactPiiDefault,
11638
- createTranscriptBodyRedactPiiSubDefault: () => createTranscriptBodyRedactPiiSubDefault,
11639
11796
  createTranscriptBodySentimentAnalysisDefault: () => createTranscriptBodySentimentAnalysisDefault,
11640
11797
  createTranscriptBodySpeakerLabelsDefault: () => createTranscriptBodySpeakerLabelsDefault,
11641
- createTranscriptBodySpeakersExpectedDefault: () => createTranscriptBodySpeakersExpectedDefault,
11642
- createTranscriptBodySpeechModelDefault: () => createTranscriptBodySpeechModelDefault,
11643
- createTranscriptBodySpeechThresholdDefault: () => createTranscriptBodySpeechThresholdDefault,
11644
- createTranscriptBodySpeechThresholdMax: () => createTranscriptBodySpeechThresholdMax,
11645
- createTranscriptBodySpeechThresholdMin: () => createTranscriptBodySpeechThresholdMin,
11798
+ createTranscriptBodySpeakerOptionsMinSpeakersExpectedDefault: () => createTranscriptBodySpeakerOptionsMinSpeakersExpectedDefault,
11799
+ createTranscriptBodySpeechUnderstandingRequestTranslationFormalDefault: () => createTranscriptBodySpeechUnderstandingRequestTranslationFormalDefault,
11800
+ createTranscriptBodySpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault: () => createTranscriptBodySpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault,
11646
11801
  createTranscriptBodySummarizationDefault: () => createTranscriptBodySummarizationDefault,
11647
- createTranscriptBodyWebhookAuthHeaderNameDefault: () => createTranscriptBodyWebhookAuthHeaderNameDefault,
11648
- createTranscriptBodyWebhookAuthHeaderValueDefault: () => createTranscriptBodyWebhookAuthHeaderValueDefault,
11802
+ createTranscriptBodyTemperatureDefault: () => createTranscriptBodyTemperatureDefault,
11649
11803
  createTranscriptResponse: () => createTranscriptResponse,
11650
- createTranscriptResponseAutoHighlightsResultResultsItemRankMax: () => createTranscriptResponseAutoHighlightsResultResultsItemRankMax,
11651
- createTranscriptResponseAutoHighlightsResultResultsItemRankMin: () => createTranscriptResponseAutoHighlightsResultResultsItemRankMin,
11652
- createTranscriptResponseConfidenceMax: () => createTranscriptResponseConfidenceMax,
11653
- createTranscriptResponseConfidenceMin: () => createTranscriptResponseConfidenceMin,
11654
- createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax: () => createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax,
11655
- createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin: () => createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin,
11656
- createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax: () => createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax,
11657
- createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin: () => createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin,
11658
- createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax: () => createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax,
11659
- createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin: () => createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin,
11660
- createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax: () => createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax,
11661
- createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin: () => createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin,
11662
- createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax: () => createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax,
11663
- createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin: () => createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin,
11664
- createTranscriptResponseContentSafetyLabelsSummaryMaxOne: () => createTranscriptResponseContentSafetyLabelsSummaryMaxOne,
11665
- createTranscriptResponseContentSafetyLabelsSummaryMinOne: () => createTranscriptResponseContentSafetyLabelsSummaryMinOne,
11666
- createTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax: () => createTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax,
11667
- createTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin: () => createTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin,
11668
- createTranscriptResponseIabCategoriesResultSummaryMaxOne: () => createTranscriptResponseIabCategoriesResultSummaryMaxOne,
11669
- createTranscriptResponseIabCategoriesResultSummaryMinOne: () => createTranscriptResponseIabCategoriesResultSummaryMinOne,
11670
- createTranscriptResponseLanguageConfidenceMax: () => createTranscriptResponseLanguageConfidenceMax,
11671
- createTranscriptResponseLanguageConfidenceMin: () => createTranscriptResponseLanguageConfidenceMin,
11672
- createTranscriptResponseLanguageConfidenceThresholdMax: () => createTranscriptResponseLanguageConfidenceThresholdMax,
11673
- createTranscriptResponseLanguageConfidenceThresholdMin: () => createTranscriptResponseLanguageConfidenceThresholdMin,
11674
- createTranscriptResponseSentimentAnalysisResultsItemConfidenceMax: () => createTranscriptResponseSentimentAnalysisResultsItemConfidenceMax,
11675
- createTranscriptResponseSentimentAnalysisResultsItemConfidenceMin: () => createTranscriptResponseSentimentAnalysisResultsItemConfidenceMin,
11676
- createTranscriptResponseSpeechModelDefault: () => createTranscriptResponseSpeechModelDefault,
11677
- createTranscriptResponseSpeechThresholdMax: () => createTranscriptResponseSpeechThresholdMax,
11678
- createTranscriptResponseSpeechThresholdMin: () => createTranscriptResponseSpeechThresholdMin,
11679
- createTranscriptResponseUtterancesItemConfidenceMax: () => createTranscriptResponseUtterancesItemConfidenceMax,
11680
- createTranscriptResponseUtterancesItemConfidenceMin: () => createTranscriptResponseUtterancesItemConfidenceMin,
11681
- createTranscriptResponseUtterancesItemWordsItemConfidenceMax: () => createTranscriptResponseUtterancesItemWordsItemConfidenceMax,
11682
- createTranscriptResponseUtterancesItemWordsItemConfidenceMin: () => createTranscriptResponseUtterancesItemWordsItemConfidenceMin,
11683
- createTranscriptResponseWordsItemConfidenceMax: () => createTranscriptResponseWordsItemConfidenceMax,
11684
- createTranscriptResponseWordsItemConfidenceMin: () => createTranscriptResponseWordsItemConfidenceMin,
11804
+ createTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault: () => createTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault,
11805
+ createTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault: () => createTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault,
11806
+ createTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault: () => createTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault,
11807
+ createTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault: () => createTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault,
11808
+ createTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault: () => createTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault,
11809
+ createTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault: () => createTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault,
11685
11810
  deleteTranscriptParams: () => deleteTranscriptParams,
11686
11811
  deleteTranscriptResponse: () => deleteTranscriptResponse,
11687
- deleteTranscriptResponseAutoHighlightsResultResultsItemRankMax: () => deleteTranscriptResponseAutoHighlightsResultResultsItemRankMax,
11688
- deleteTranscriptResponseAutoHighlightsResultResultsItemRankMin: () => deleteTranscriptResponseAutoHighlightsResultResultsItemRankMin,
11689
- deleteTranscriptResponseConfidenceMax: () => deleteTranscriptResponseConfidenceMax,
11690
- deleteTranscriptResponseConfidenceMin: () => deleteTranscriptResponseConfidenceMin,
11691
- deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax: () => deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax,
11692
- deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin: () => deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin,
11693
- deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax: () => deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax,
11694
- deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin: () => deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin,
11695
- deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax: () => deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax,
11696
- deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin: () => deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin,
11697
- deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax: () => deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax,
11698
- deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin: () => deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin,
11699
- deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax: () => deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax,
11700
- deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin: () => deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin,
11701
- deleteTranscriptResponseContentSafetyLabelsSummaryMaxOne: () => deleteTranscriptResponseContentSafetyLabelsSummaryMaxOne,
11702
- deleteTranscriptResponseContentSafetyLabelsSummaryMinOne: () => deleteTranscriptResponseContentSafetyLabelsSummaryMinOne,
11703
- deleteTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax: () => deleteTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax,
11704
- deleteTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin: () => deleteTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin,
11705
- deleteTranscriptResponseIabCategoriesResultSummaryMaxOne: () => deleteTranscriptResponseIabCategoriesResultSummaryMaxOne,
11706
- deleteTranscriptResponseIabCategoriesResultSummaryMinOne: () => deleteTranscriptResponseIabCategoriesResultSummaryMinOne,
11707
- deleteTranscriptResponseLanguageConfidenceMax: () => deleteTranscriptResponseLanguageConfidenceMax,
11708
- deleteTranscriptResponseLanguageConfidenceMin: () => deleteTranscriptResponseLanguageConfidenceMin,
11709
- deleteTranscriptResponseLanguageConfidenceThresholdMax: () => deleteTranscriptResponseLanguageConfidenceThresholdMax,
11710
- deleteTranscriptResponseLanguageConfidenceThresholdMin: () => deleteTranscriptResponseLanguageConfidenceThresholdMin,
11711
- deleteTranscriptResponseSentimentAnalysisResultsItemConfidenceMax: () => deleteTranscriptResponseSentimentAnalysisResultsItemConfidenceMax,
11712
- deleteTranscriptResponseSentimentAnalysisResultsItemConfidenceMin: () => deleteTranscriptResponseSentimentAnalysisResultsItemConfidenceMin,
11713
- deleteTranscriptResponseSpeechModelDefault: () => deleteTranscriptResponseSpeechModelDefault,
11714
- deleteTranscriptResponseSpeechThresholdMax: () => deleteTranscriptResponseSpeechThresholdMax,
11715
- deleteTranscriptResponseSpeechThresholdMin: () => deleteTranscriptResponseSpeechThresholdMin,
11716
- deleteTranscriptResponseUtterancesItemConfidenceMax: () => deleteTranscriptResponseUtterancesItemConfidenceMax,
11717
- deleteTranscriptResponseUtterancesItemConfidenceMin: () => deleteTranscriptResponseUtterancesItemConfidenceMin,
11718
- deleteTranscriptResponseUtterancesItemWordsItemConfidenceMax: () => deleteTranscriptResponseUtterancesItemWordsItemConfidenceMax,
11719
- deleteTranscriptResponseUtterancesItemWordsItemConfidenceMin: () => deleteTranscriptResponseUtterancesItemWordsItemConfidenceMin,
11720
- deleteTranscriptResponseWordsItemConfidenceMax: () => deleteTranscriptResponseWordsItemConfidenceMax,
11721
- deleteTranscriptResponseWordsItemConfidenceMin: () => deleteTranscriptResponseWordsItemConfidenceMin,
11722
- getLemurResponseParams: () => getLemurResponseParams,
11723
- getLemurResponseResponse: () => getLemurResponseResponse,
11724
- getLemurResponseResponseUsageInputTokensMin: () => getLemurResponseResponseUsageInputTokensMin,
11725
- getLemurResponseResponseUsageInputTokensMinOne: () => getLemurResponseResponseUsageInputTokensMinOne,
11726
- getLemurResponseResponseUsageOutputTokensMin: () => getLemurResponseResponseUsageOutputTokensMin,
11727
- getLemurResponseResponseUsageOutputTokensMinOne: () => getLemurResponseResponseUsageOutputTokensMinOne,
11812
+ deleteTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault: () => deleteTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault,
11813
+ deleteTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault: () => deleteTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault,
11814
+ deleteTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault: () => deleteTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault,
11815
+ deleteTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault: () => deleteTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault,
11816
+ deleteTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault: () => deleteTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault,
11817
+ deleteTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault: () => deleteTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault,
11728
11818
  getRedactedAudioParams: () => getRedactedAudioParams,
11729
11819
  getRedactedAudioResponse: () => getRedactedAudioResponse,
11730
11820
  getSubtitlesParams: () => getSubtitlesParams,
11731
11821
  getSubtitlesQueryParams: () => getSubtitlesQueryParams,
11822
+ getSubtitlesResponse: () => getSubtitlesResponse,
11732
11823
  getTranscriptParagraphsParams: () => getTranscriptParagraphsParams,
11733
11824
  getTranscriptParagraphsResponse: () => getTranscriptParagraphsResponse,
11734
- getTranscriptParagraphsResponseConfidenceMax: () => getTranscriptParagraphsResponseConfidenceMax,
11735
- getTranscriptParagraphsResponseConfidenceMin: () => getTranscriptParagraphsResponseConfidenceMin,
11736
- getTranscriptParagraphsResponseParagraphsItemConfidenceMax: () => getTranscriptParagraphsResponseParagraphsItemConfidenceMax,
11737
- getTranscriptParagraphsResponseParagraphsItemConfidenceMin: () => getTranscriptParagraphsResponseParagraphsItemConfidenceMin,
11738
- getTranscriptParagraphsResponseParagraphsItemWordsItemConfidenceMax: () => getTranscriptParagraphsResponseParagraphsItemWordsItemConfidenceMax,
11739
- getTranscriptParagraphsResponseParagraphsItemWordsItemConfidenceMin: () => getTranscriptParagraphsResponseParagraphsItemWordsItemConfidenceMin,
11740
11825
  getTranscriptParams: () => getTranscriptParams,
11741
11826
  getTranscriptResponse: () => getTranscriptResponse,
11742
- getTranscriptResponseAutoHighlightsResultResultsItemRankMax: () => getTranscriptResponseAutoHighlightsResultResultsItemRankMax,
11743
- getTranscriptResponseAutoHighlightsResultResultsItemRankMin: () => getTranscriptResponseAutoHighlightsResultResultsItemRankMin,
11744
- getTranscriptResponseConfidenceMax: () => getTranscriptResponseConfidenceMax,
11745
- getTranscriptResponseConfidenceMin: () => getTranscriptResponseConfidenceMin,
11746
- getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax: () => getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax,
11747
- getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin: () => getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin,
11748
- getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax: () => getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax,
11749
- getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin: () => getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin,
11750
- getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax: () => getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax,
11751
- getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin: () => getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin,
11752
- getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax: () => getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax,
11753
- getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin: () => getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin,
11754
- getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax: () => getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax,
11755
- getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin: () => getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin,
11756
- getTranscriptResponseContentSafetyLabelsSummaryMaxOne: () => getTranscriptResponseContentSafetyLabelsSummaryMaxOne,
11757
- getTranscriptResponseContentSafetyLabelsSummaryMinOne: () => getTranscriptResponseContentSafetyLabelsSummaryMinOne,
11758
- getTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax: () => getTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax,
11759
- getTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin: () => getTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin,
11760
- getTranscriptResponseIabCategoriesResultSummaryMaxOne: () => getTranscriptResponseIabCategoriesResultSummaryMaxOne,
11761
- getTranscriptResponseIabCategoriesResultSummaryMinOne: () => getTranscriptResponseIabCategoriesResultSummaryMinOne,
11762
- getTranscriptResponseLanguageConfidenceMax: () => getTranscriptResponseLanguageConfidenceMax,
11763
- getTranscriptResponseLanguageConfidenceMin: () => getTranscriptResponseLanguageConfidenceMin,
11764
- getTranscriptResponseLanguageConfidenceThresholdMax: () => getTranscriptResponseLanguageConfidenceThresholdMax,
11765
- getTranscriptResponseLanguageConfidenceThresholdMin: () => getTranscriptResponseLanguageConfidenceThresholdMin,
11766
- getTranscriptResponseSentimentAnalysisResultsItemConfidenceMax: () => getTranscriptResponseSentimentAnalysisResultsItemConfidenceMax,
11767
- getTranscriptResponseSentimentAnalysisResultsItemConfidenceMin: () => getTranscriptResponseSentimentAnalysisResultsItemConfidenceMin,
11768
- getTranscriptResponseSpeechModelDefault: () => getTranscriptResponseSpeechModelDefault,
11769
- getTranscriptResponseSpeechThresholdMax: () => getTranscriptResponseSpeechThresholdMax,
11770
- getTranscriptResponseSpeechThresholdMin: () => getTranscriptResponseSpeechThresholdMin,
11771
- getTranscriptResponseUtterancesItemConfidenceMax: () => getTranscriptResponseUtterancesItemConfidenceMax,
11772
- getTranscriptResponseUtterancesItemConfidenceMin: () => getTranscriptResponseUtterancesItemConfidenceMin,
11773
- getTranscriptResponseUtterancesItemWordsItemConfidenceMax: () => getTranscriptResponseUtterancesItemWordsItemConfidenceMax,
11774
- getTranscriptResponseUtterancesItemWordsItemConfidenceMin: () => getTranscriptResponseUtterancesItemWordsItemConfidenceMin,
11775
- getTranscriptResponseWordsItemConfidenceMax: () => getTranscriptResponseWordsItemConfidenceMax,
11776
- getTranscriptResponseWordsItemConfidenceMin: () => getTranscriptResponseWordsItemConfidenceMin,
11827
+ getTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault: () => getTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault,
11828
+ getTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault: () => getTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault,
11829
+ getTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault: () => getTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault,
11830
+ getTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault: () => getTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault,
11831
+ getTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault: () => getTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault,
11832
+ getTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault: () => getTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault,
11777
11833
  getTranscriptSentencesParams: () => getTranscriptSentencesParams,
11778
11834
  getTranscriptSentencesResponse: () => getTranscriptSentencesResponse,
11779
- getTranscriptSentencesResponseConfidenceMax: () => getTranscriptSentencesResponseConfidenceMax,
11780
- getTranscriptSentencesResponseConfidenceMin: () => getTranscriptSentencesResponseConfidenceMin,
11781
- getTranscriptSentencesResponseSentencesItemConfidenceMax: () => getTranscriptSentencesResponseSentencesItemConfidenceMax,
11782
- getTranscriptSentencesResponseSentencesItemConfidenceMin: () => getTranscriptSentencesResponseSentencesItemConfidenceMin,
11783
- getTranscriptSentencesResponseSentencesItemWordsItemConfidenceMax: () => getTranscriptSentencesResponseSentencesItemWordsItemConfidenceMax,
11784
- getTranscriptSentencesResponseSentencesItemWordsItemConfidenceMin: () => getTranscriptSentencesResponseSentencesItemWordsItemConfidenceMin,
11785
- lemurQuestionAnswerBody: () => lemurQuestionAnswerBody,
11786
- lemurQuestionAnswerBodyFinalModelDefault: () => lemurQuestionAnswerBodyFinalModelDefault,
11787
- lemurQuestionAnswerBodyMaxOutputSizeDefault: () => lemurQuestionAnswerBodyMaxOutputSizeDefault,
11788
- lemurQuestionAnswerBodyTemperatureDefault: () => lemurQuestionAnswerBodyTemperatureDefault,
11789
- lemurQuestionAnswerBodyTemperatureMax: () => lemurQuestionAnswerBodyTemperatureMax,
11790
- lemurQuestionAnswerBodyTemperatureMin: () => lemurQuestionAnswerBodyTemperatureMin,
11791
- lemurQuestionAnswerResponse: () => lemurQuestionAnswerResponse,
11792
- lemurQuestionAnswerResponseUsageInputTokensMin: () => lemurQuestionAnswerResponseUsageInputTokensMin,
11793
- lemurQuestionAnswerResponseUsageOutputTokensMin: () => lemurQuestionAnswerResponseUsageOutputTokensMin,
11794
- lemurSummaryBody: () => lemurSummaryBody,
11795
- lemurSummaryBodyFinalModelDefault: () => lemurSummaryBodyFinalModelDefault,
11796
- lemurSummaryBodyMaxOutputSizeDefault: () => lemurSummaryBodyMaxOutputSizeDefault,
11797
- lemurSummaryBodyTemperatureDefault: () => lemurSummaryBodyTemperatureDefault,
11798
- lemurSummaryBodyTemperatureMax: () => lemurSummaryBodyTemperatureMax,
11799
- lemurSummaryBodyTemperatureMin: () => lemurSummaryBodyTemperatureMin,
11800
- lemurSummaryResponse: () => lemurSummaryResponse,
11801
- lemurSummaryResponseUsageInputTokensMin: () => lemurSummaryResponseUsageInputTokensMin,
11802
- lemurSummaryResponseUsageOutputTokensMin: () => lemurSummaryResponseUsageOutputTokensMin,
11803
- lemurTaskBody: () => lemurTaskBody,
11804
- lemurTaskBodyFinalModelDefault: () => lemurTaskBodyFinalModelDefault,
11805
- lemurTaskBodyMaxOutputSizeDefault: () => lemurTaskBodyMaxOutputSizeDefault,
11806
- lemurTaskBodyTemperatureDefault: () => lemurTaskBodyTemperatureDefault,
11807
- lemurTaskBodyTemperatureMax: () => lemurTaskBodyTemperatureMax,
11808
- lemurTaskBodyTemperatureMin: () => lemurTaskBodyTemperatureMin,
11809
- lemurTaskResponse: () => lemurTaskResponse,
11810
- lemurTaskResponseUsageInputTokensMin: () => lemurTaskResponseUsageInputTokensMin,
11811
- lemurTaskResponseUsageOutputTokensMin: () => lemurTaskResponseUsageOutputTokensMin,
11812
11835
  listTranscriptsQueryLimitDefault: () => listTranscriptsQueryLimitDefault,
11813
- listTranscriptsQueryLimitMax: () => listTranscriptsQueryLimitMax,
11814
11836
  listTranscriptsQueryParams: () => listTranscriptsQueryParams,
11815
11837
  listTranscriptsQueryThrottledOnlyDefault: () => listTranscriptsQueryThrottledOnlyDefault,
11816
11838
  listTranscriptsResponse: () => listTranscriptsResponse,
11817
- listTranscriptsResponseTranscriptsItemCompletedRegExp: () => listTranscriptsResponseTranscriptsItemCompletedRegExp,
11818
- listTranscriptsResponseTranscriptsItemCreatedRegExp: () => listTranscriptsResponseTranscriptsItemCreatedRegExp,
11819
- purgeLemurRequestDataParams: () => purgeLemurRequestDataParams,
11820
- purgeLemurRequestDataResponse: () => purgeLemurRequestDataResponse,
11821
11839
  uploadFileResponse: () => uploadFileResponse,
11822
11840
  wordSearchParams: () => wordSearchParams,
11823
11841
  wordSearchQueryParams: () => wordSearchQueryParams,
@@ -11827,43 +11845,192 @@ var import_zod3 = require("zod");
11827
11845
  var uploadFileResponse = import_zod3.z.object({
11828
11846
  upload_url: import_zod3.z.string().describe("A URL that points to your audio file, accessible only by AssemblyAI's servers\n")
11829
11847
  });
11830
- var createTranscriptBodyLanguageCodeDefault = "en_us";
11831
- var createTranscriptBodyLanguageDetectionDefault = false;
11832
- var createTranscriptBodyLanguageConfidenceThresholdDefault = 0;
11833
- var createTranscriptBodyLanguageConfidenceThresholdMin = 0;
11834
- var createTranscriptBodyLanguageConfidenceThresholdMax = 1;
11835
- var createTranscriptBodySpeechModelDefault = "best";
11836
- var createTranscriptBodyPunctuateDefault = true;
11837
- var createTranscriptBodyFormatTextDefault = true;
11838
- var createTranscriptBodyDisfluenciesDefault = false;
11839
- var createTranscriptBodyMultichannelDefault = false;
11840
- var createTranscriptBodyWebhookAuthHeaderNameDefault = null;
11841
- var createTranscriptBodyWebhookAuthHeaderValueDefault = null;
11848
+ var createTranscriptBodyAutoChaptersDefault = false;
11842
11849
  var createTranscriptBodyAutoHighlightsDefault = false;
11843
- var createTranscriptBodyFilterProfanityDefault = false;
11844
- var createTranscriptBodyRedactPiiDefault = false;
11845
- var createTranscriptBodyRedactPiiAudioDefault = false;
11846
- var createTranscriptBodyRedactPiiSubDefault = "hash";
11847
- var createTranscriptBodySpeakerLabelsDefault = false;
11848
- var createTranscriptBodySpeakersExpectedDefault = null;
11849
11850
  var createTranscriptBodyContentSafetyDefault = false;
11850
11851
  var createTranscriptBodyContentSafetyConfidenceDefault = 50;
11851
- var createTranscriptBodyContentSafetyConfidenceMin = 25;
11852
- var createTranscriptBodyContentSafetyConfidenceMax = 100;
11852
+ var createTranscriptBodyDisfluenciesDefault = false;
11853
+ var createTranscriptBodyEntityDetectionDefault = false;
11854
+ var createTranscriptBodyFilterProfanityDefault = false;
11855
+ var createTranscriptBodyFormatTextDefault = true;
11853
11856
  var createTranscriptBodyIabCategoriesDefault = false;
11857
+ var createTranscriptBodyLanguageDetectionDefault = false;
11858
+ var createTranscriptBodyLanguageDetectionOptionsFallbackLanguageDefault = "auto";
11859
+ var createTranscriptBodyLanguageDetectionOptionsCodeSwitchingDefault = false;
11860
+ var createTranscriptBodyLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault = 0.3;
11861
+ var createTranscriptBodyMultichannelDefault = false;
11862
+ var createTranscriptBodyPunctuateDefault = true;
11863
+ var createTranscriptBodyRedactPiiDefault = false;
11864
+ var createTranscriptBodyRedactPiiAudioDefault = false;
11865
+ var createTranscriptBodyRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault = false;
11854
11866
  var createTranscriptBodySentimentAnalysisDefault = false;
11855
- var createTranscriptBodyAutoChaptersDefault = false;
11856
- var createTranscriptBodyEntityDetectionDefault = false;
11857
- var createTranscriptBodySpeechThresholdDefault = 0;
11858
- var createTranscriptBodySpeechThresholdMin = 0;
11859
- var createTranscriptBodySpeechThresholdMax = 1;
11867
+ var createTranscriptBodySpeakerLabelsDefault = false;
11868
+ var createTranscriptBodySpeakerOptionsMinSpeakersExpectedDefault = 1;
11869
+ var createTranscriptBodySpeechUnderstandingRequestTranslationFormalDefault = true;
11870
+ var createTranscriptBodySpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault = false;
11860
11871
  var createTranscriptBodySummarizationDefault = false;
11872
+ var createTranscriptBodyTemperatureDefault = 0;
11861
11873
  var createTranscriptBodyCustomTopicsDefault = false;
11862
11874
  var createTranscriptBody = import_zod3.z.object({
11863
- audio_url: import_zod3.z.string().describe("The URL of the audio or video file to transcribe.")
11864
- }).and(
11865
- import_zod3.z.object({
11866
- language_code: import_zod3.z.enum([
11875
+ audio_end_at: import_zod3.z.number().optional().describe(
11876
+ "The point in time, in milliseconds, to stop transcribing in your media file. See [Set the start and end of the transcript](https://www.assemblyai.com/docs/pre-recorded-audio/set-the-start-and-end-of-the-transcript) for more details."
11877
+ ),
11878
+ audio_start_from: import_zod3.z.number().optional().describe(
11879
+ "The point in time, in milliseconds, to begin transcribing in your media file. See [Set the start and end of the transcript](https://www.assemblyai.com/docs/pre-recorded-audio/set-the-start-and-end-of-the-transcript) for more details."
11880
+ ),
11881
+ auto_chapters: import_zod3.z.boolean().optional().describe(
11882
+ "Enable [Auto Chapters](https://www.assemblyai.com/docs/speech-understanding/auto-chapters), can be true or false. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible chapter summaries. See the [updated Auto Chapters page](https://www.assemblyai.com/docs/speech-understanding/auto-chapters) for details.\n\nNote: This parameter is only supported for the Universal-2 model.\n"
11883
+ ),
11884
+ auto_highlights: import_zod3.z.boolean().optional().describe(
11885
+ "Enable [Key Phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases), either true or false"
11886
+ ),
11887
+ content_safety: import_zod3.z.boolean().optional().describe(
11888
+ "Enable [Content Moderation](https://www.assemblyai.com/docs/content-moderation), can be true or false"
11889
+ ),
11890
+ content_safety_confidence: import_zod3.z.number().default(createTranscriptBodyContentSafetyConfidenceDefault).describe(
11891
+ "The confidence threshold for the [Content Moderation](https://www.assemblyai.com/docs/content-moderation) model. Values must be between 25 and 100."
11892
+ ),
11893
+ custom_spelling: import_zod3.z.array(
11894
+ import_zod3.z.object({
11895
+ from: import_zod3.z.array(import_zod3.z.string()).describe("Words or phrases to replace"),
11896
+ to: import_zod3.z.string().describe("Word to replace with")
11897
+ }).describe(
11898
+ "Object containing words or phrases to replace, and the word or phrase to replace with"
11899
+ )
11900
+ ).optional().describe(
11901
+ "Customize how words are spelled and formatted using to and from values. See [Custom Spelling](https://www.assemblyai.com/docs/pre-recorded-audio/custom-spelling) for more details."
11902
+ ),
11903
+ disfluencies: import_zod3.z.boolean().optional().describe(
11904
+ 'Transcribe [Filler Words](https://www.assemblyai.com/docs/pre-recorded-audio/filler-words), like "umm", in your media file; can be true or false'
11905
+ ),
11906
+ domain: import_zod3.z.string().nullish().describe(
11907
+ 'Enable domain-specific transcription models to improve accuracy for specialized terminology. Set to `"medical-v1"` to enable [Medical Mode](https://www.assemblyai.com/docs/pre-recorded-audio/medical-mode) for improved accuracy of medical terms such as medications, procedures, conditions, and dosages.\n\nSupported languages: English (`en`), Spanish (`es`), German (`de`), French (`fr`). If used with an unsupported language, the parameter is ignored and a warning is returned.\n'
11908
+ ),
11909
+ entity_detection: import_zod3.z.boolean().optional().describe(
11910
+ "Enable [Entity Detection](https://www.assemblyai.com/docs/speech-understanding/entity-detection), can be true or false"
11911
+ ),
11912
+ filter_profanity: import_zod3.z.boolean().optional().describe(
11913
+ "Filter profanity from the transcribed text, can be true or false. See [Profanity Filtering](https://www.assemblyai.com/docs/profanity-filtering) for more details."
11914
+ ),
11915
+ format_text: import_zod3.z.boolean().default(createTranscriptBodyFormatTextDefault).describe(
11916
+ "Enable [Text Formatting](https://www.assemblyai.com/docs/pre-recorded-audio), can be true or false"
11917
+ ),
11918
+ iab_categories: import_zod3.z.boolean().optional().describe(
11919
+ "Enable [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection), can be true or false"
11920
+ ),
11921
+ keyterms_prompt: import_zod3.z.array(import_zod3.z.string()).optional().describe(
11922
+ "Improve accuracy with up to 200 (for Universal-2) or 1000 (for Universal-3 Pro) domain-specific words or phrases (maximum 6 words per phrase). See [Keyterms Prompting](https://www.assemblyai.com/docs/pre-recorded-audio/keyterms-prompting) for more details.\n"
11923
+ ),
11924
+ language_code: import_zod3.z.enum([
11925
+ "en",
11926
+ "en_au",
11927
+ "en_uk",
11928
+ "en_us",
11929
+ "es",
11930
+ "fr",
11931
+ "de",
11932
+ "it",
11933
+ "pt",
11934
+ "nl",
11935
+ "af",
11936
+ "sq",
11937
+ "am",
11938
+ "ar",
11939
+ "hy",
11940
+ "as",
11941
+ "az",
11942
+ "ba",
11943
+ "eu",
11944
+ "be",
11945
+ "bn",
11946
+ "bs",
11947
+ "br",
11948
+ "bg",
11949
+ "my",
11950
+ "ca",
11951
+ "zh",
11952
+ "hr",
11953
+ "cs",
11954
+ "da",
11955
+ "et",
11956
+ "fo",
11957
+ "fi",
11958
+ "gl",
11959
+ "ka",
11960
+ "el",
11961
+ "gu",
11962
+ "ht",
11963
+ "ha",
11964
+ "haw",
11965
+ "he",
11966
+ "hi",
11967
+ "hu",
11968
+ "is",
11969
+ "id",
11970
+ "ja",
11971
+ "jw",
11972
+ "kn",
11973
+ "kk",
11974
+ "km",
11975
+ "ko",
11976
+ "lo",
11977
+ "la",
11978
+ "lv",
11979
+ "ln",
11980
+ "lt",
11981
+ "lb",
11982
+ "mk",
11983
+ "mg",
11984
+ "ms",
11985
+ "ml",
11986
+ "mt",
11987
+ "mi",
11988
+ "mr",
11989
+ "mn",
11990
+ "ne",
11991
+ "no",
11992
+ "nn",
11993
+ "oc",
11994
+ "pa",
11995
+ "ps",
11996
+ "fa",
11997
+ "pl",
11998
+ "ro",
11999
+ "ru",
12000
+ "sa",
12001
+ "sr",
12002
+ "sn",
12003
+ "sd",
12004
+ "si",
12005
+ "sk",
12006
+ "sl",
12007
+ "so",
12008
+ "su",
12009
+ "sw",
12010
+ "sv",
12011
+ "tl",
12012
+ "tg",
12013
+ "ta",
12014
+ "tt",
12015
+ "te",
12016
+ "th",
12017
+ "bo",
12018
+ "tr",
12019
+ "tk",
12020
+ "uk",
12021
+ "ur",
12022
+ "uz",
12023
+ "vi",
12024
+ "cy",
12025
+ "yi",
12026
+ "yo"
12027
+ ]).describe(
12028
+ "The language of your audio file. Possible values are found in [Supported Languages](https://www.assemblyai.com/docs/pre-recorded-audio/supported-languages).\nThe default value is 'en_us'.\n"
12029
+ ).or(import_zod3.z.null()).optional().describe(
12030
+ "The language of your audio file. Possible values are found in [Supported Languages](https://www.assemblyai.com/docs/pre-recorded-audio/supported-languages).\nThe default value is 'en_us'.\n"
12031
+ ),
12032
+ language_codes: import_zod3.z.array(
12033
+ import_zod3.z.enum([
11867
12034
  "en",
11868
12035
  "en_au",
11869
12036
  "en_uk",
@@ -11967,53 +12134,365 @@ var createTranscriptBody = import_zod3.z.object({
11967
12134
  "yi",
11968
12135
  "yo"
11969
12136
  ]).describe(
11970
- "The language of your audio file. Possible values are found in [Supported Languages](https://www.assemblyai.com/docs/concepts/supported-languages).\nThe default value is 'en_us'.\n"
11971
- ).or(import_zod3.z.string()).or(import_zod3.z.null()).default(createTranscriptBodyLanguageCodeDefault).describe(
11972
- "The language of your audio file. Possible values are found in [Supported Languages](https://www.assemblyai.com/docs/concepts/supported-languages).\nThe default value is 'en_us'.\n"
11973
- ),
11974
- language_detection: import_zod3.z.boolean().optional().describe(
11975
- "Enable [Automatic language detection](https://www.assemblyai.com/docs/models/speech-recognition#automatic-language-detection), either true or false."
11976
- ),
11977
- language_confidence_threshold: import_zod3.z.number().min(createTranscriptBodyLanguageConfidenceThresholdMin).max(createTranscriptBodyLanguageConfidenceThresholdMax).optional().describe(
11978
- "The confidence threshold for the automatically detected language.\nAn error will be returned if the language confidence is below this threshold.\nDefaults to 0.\n"
11979
- ),
11980
- speech_model: import_zod3.z.enum(["best", "slam-1", "universal"]).describe("The speech model to use for the transcription.").or(import_zod3.z.null()).default(createTranscriptBodySpeechModelDefault).describe(
11981
- 'The speech model to use for the transcription. When `null`, the "best" model is used.'
11982
- ),
11983
- punctuate: import_zod3.z.boolean().default(createTranscriptBodyPunctuateDefault).describe("Enable Automatic Punctuation, can be true or false"),
11984
- format_text: import_zod3.z.boolean().default(createTranscriptBodyFormatTextDefault).describe("Enable Text Formatting, can be true or false"),
11985
- disfluencies: import_zod3.z.boolean().optional().describe(
11986
- 'Transcribe Filler Words, like "umm", in your media file; can be true or false'
11987
- ),
11988
- multichannel: import_zod3.z.boolean().optional().describe(
11989
- "Enable [Multichannel](https://www.assemblyai.com/docs/models/speech-recognition#multichannel-transcription) transcription, can be true or false."
11990
- ),
11991
- webhook_url: import_zod3.z.string().optional().describe(
11992
- "The URL to which we send webhook requests.\nWe sends two different types of webhook requests.\nOne request when a transcript is completed or failed, and one request when the redacted audio is ready if redact_pii_audio is enabled.\n"
12137
+ "The language of your audio file. Possible values are found in [Supported Languages](https://www.assemblyai.com/docs/pre-recorded-audio/supported-languages).\nThe default value is 'en_us'.\n"
12138
+ )
12139
+ ).nullish().describe(
12140
+ "The language codes of your audio file. Used for [Code switching](/docs/speech-to-text/pre-recorded-audio/code-switching)\nOne of the values specified must be `en`.\n"
12141
+ ),
12142
+ language_confidence_threshold: import_zod3.z.number().optional().describe(
12143
+ "The confidence threshold for the automatically detected language.\nAn error will be returned if the language confidence is below this threshold.\nDefaults to 0. See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.\n"
12144
+ ),
12145
+ language_detection: import_zod3.z.boolean().optional().describe(
12146
+ "Enable [Automatic language detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection), either true or false."
12147
+ ),
12148
+ language_detection_options: import_zod3.z.object({
12149
+ expected_languages: import_zod3.z.array(import_zod3.z.string()).optional().describe(
12150
+ 'List of languages expected in the audio file. Defaults to `["all"]` when unspecified. See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.'
11993
12151
  ),
11994
- webhook_auth_header_name: import_zod3.z.string().nullish().describe(
11995
- "The header name to be sent with the transcript completed or failed webhook requests"
12152
+ fallback_language: import_zod3.z.string().default(createTranscriptBodyLanguageDetectionOptionsFallbackLanguageDefault).describe(
12153
+ 'If the detected language of the audio file is not in the list of expected languages, the `fallback_language` is used. Specify `["auto"]` to let our model choose the fallback language from `expected_languages` with the highest confidence score. See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.\n'
11996
12154
  ),
11997
- webhook_auth_header_value: import_zod3.z.string().nullish().describe(
11998
- "The header value to send back with the transcript completed or failed webhook requests for added security"
12155
+ code_switching: import_zod3.z.boolean().optional().describe(
12156
+ "Whether [code switching](/docs/speech-to-text/pre-recorded-audio/code-switching) should be detected.\n"
11999
12157
  ),
12000
- auto_highlights: import_zod3.z.boolean().optional().describe("Enable Key Phrases, either true or false"),
12001
- audio_start_from: import_zod3.z.number().optional().describe("The point in time, in milliseconds, to begin transcribing in your media file"),
12002
- audio_end_at: import_zod3.z.number().optional().describe("The point in time, in milliseconds, to stop transcribing in your media file"),
12003
- word_boost: import_zod3.z.array(import_zod3.z.string()).optional().describe("The list of custom vocabulary to boost transcription probability for"),
12004
- boost_param: import_zod3.z.enum(["low", "default", "high"]).optional().describe("How much to boost specified words"),
12005
- filter_profanity: import_zod3.z.boolean().optional().describe("Filter profanity from the transcribed text, can be true or false"),
12006
- redact_pii: import_zod3.z.boolean().optional().describe(
12007
- "Redact PII from the transcribed text using the Redact PII model, can be true or false"
12158
+ code_switching_confidence_threshold: import_zod3.z.number().default(
12159
+ createTranscriptBodyLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault
12160
+ ).describe(
12161
+ "The confidence threshold for [code switching](/docs/speech-to-text/pre-recorded-audio/code-switching) detection. If the code switching confidence is below this threshold, the transcript will be processed in the language with the highest `language_detection_confidence` score.\n"
12162
+ )
12163
+ }).optional().describe(
12164
+ "Specify options for [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection)."
12165
+ ),
12166
+ multichannel: import_zod3.z.boolean().optional().describe(
12167
+ "Enable [Multichannel](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) transcription, can be true or false."
12168
+ ),
12169
+ prompt: import_zod3.z.string().optional().describe(
12170
+ "Provide natural language prompting of up to 1,500 words of contextual information to the model. See the [Prompting Guide](https://www.assemblyai.com/docs/pre-recorded-audio/prompting) for best practices.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n"
12171
+ ),
12172
+ punctuate: import_zod3.z.boolean().default(createTranscriptBodyPunctuateDefault).describe(
12173
+ "Enable [Automatic Punctuation](https://www.assemblyai.com/docs/pre-recorded-audio), can be true or false"
12174
+ ),
12175
+ redact_pii: import_zod3.z.boolean().optional().describe(
12176
+ "Redact PII from the transcribed text using the Redact PII model, can be true or false. See [PII Redaction](https://www.assemblyai.com/docs/pii-redaction) for more details."
12177
+ ),
12178
+ redact_pii_audio: import_zod3.z.boolean().optional().describe(
12179
+ 'Generate a copy of the original media file with spoken PII "beeped" out, can be true or false. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more details.'
12180
+ ),
12181
+ redact_pii_audio_options: import_zod3.z.object({
12182
+ return_redacted_no_speech_audio: import_zod3.z.boolean().optional().describe(
12183
+ "By default, audio redaction provides redacted audio URLs only when speech is detected. However, if your use-case specifically requires redacted audio files even for silent audio files without any dialogue, you can opt to receive these URLs by setting this parameter to `true`."
12008
12184
  ),
12009
- redact_pii_audio: import_zod3.z.boolean().optional().describe(
12010
- 'Generate a copy of the original media file with spoken PII "beeped" out, can be true or false. See [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more details.'
12185
+ override_audio_redaction_method: import_zod3.z.enum(["silence"]).optional().describe(
12186
+ "Specify the method used to redact audio. By default, redacted audio uses a beep sound. Set to `silence` to replace PII with silence instead of a beep."
12187
+ )
12188
+ }).optional().describe(
12189
+ "Specify options for [PII redacted audio](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) files."
12190
+ ),
12191
+ redact_pii_audio_quality: import_zod3.z.enum(["mp3", "wav"]).optional().describe(
12192
+ "Controls the filetype of the audio created by redact_pii_audio. Currently supports mp3 (default) and wav. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more details."
12193
+ ),
12194
+ redact_pii_policies: import_zod3.z.array(
12195
+ import_zod3.z.enum([
12196
+ "account_number",
12197
+ "banking_information",
12198
+ "blood_type",
12199
+ "credit_card_cvv",
12200
+ "credit_card_expiration",
12201
+ "credit_card_number",
12202
+ "date",
12203
+ "date_interval",
12204
+ "date_of_birth",
12205
+ "drivers_license",
12206
+ "drug",
12207
+ "duration",
12208
+ "email_address",
12209
+ "event",
12210
+ "filename",
12211
+ "gender_sexuality",
12212
+ "healthcare_number",
12213
+ "injury",
12214
+ "ip_address",
12215
+ "language",
12216
+ "location",
12217
+ "marital_status",
12218
+ "medical_condition",
12219
+ "medical_process",
12220
+ "money_amount",
12221
+ "nationality",
12222
+ "number_sequence",
12223
+ "occupation",
12224
+ "organization",
12225
+ "passport_number",
12226
+ "password",
12227
+ "person_age",
12228
+ "person_name",
12229
+ "phone_number",
12230
+ "physical_attribute",
12231
+ "political_affiliation",
12232
+ "religion",
12233
+ "statistics",
12234
+ "time",
12235
+ "url",
12236
+ "us_social_security_number",
12237
+ "username",
12238
+ "vehicle_id",
12239
+ "zodiac_sign"
12240
+ ]).describe("The type of PII to redact")
12241
+ ).optional().describe(
12242
+ "The list of PII Redaction policies to enable. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction) for more details."
12243
+ ),
12244
+ redact_pii_sub: import_zod3.z.enum(["entity_name", "hash"]).describe(
12245
+ "The replacement logic for detected PII, can be `entity_name` or `hash`. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction) for more details."
12246
+ ).or(import_zod3.z.null()).optional().describe(
12247
+ "The replacement logic for detected PII, can be `entity_type` or `hash`. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction) for more details."
12248
+ ),
12249
+ sentiment_analysis: import_zod3.z.boolean().optional().describe(
12250
+ "Enable [Sentiment Analysis](https://www.assemblyai.com/docs/speech-understanding/sentiment-analysis), can be true or false"
12251
+ ),
12252
+ speaker_labels: import_zod3.z.boolean().optional().describe(
12253
+ "Enable [Speaker diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization), can be true or false"
12254
+ ),
12255
+ speaker_options: import_zod3.z.object({
12256
+ min_speakers_expected: import_zod3.z.number().default(createTranscriptBodySpeakerOptionsMinSpeakersExpectedDefault).describe(
12257
+ "The minimum number of speakers expected in the audio file. See [Set a range of possible speakers](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization#set-a-range-of-possible-speakers) for more details."
12011
12258
  ),
12012
- redact_pii_audio_quality: import_zod3.z.enum(["mp3", "wav"]).optional().describe(
12013
- "Controls the filetype of the audio created by redact_pii_audio. Currently supports mp3 (default) and wav. See [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more details."
12259
+ max_speakers_expected: import_zod3.z.number().optional().describe(
12260
+ "<Warning>Setting this parameter too high may hurt model accuracy</Warning>\nThe maximum number of speakers expected in the audio file. The default depends on audio duration: no limit for 0-2 minutes, 10 for 2-10 minutes, and 30 for 10+ minutes. See [Set a range of possible speakers](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization#set-a-range-of-possible-speakers) for more details.\n"
12261
+ )
12262
+ }).optional().describe(
12263
+ "Specify options for [Speaker diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization#set-a-range-of-possible-speakers). Use this to set a range of possible speakers."
12264
+ ),
12265
+ speakers_expected: import_zod3.z.number().nullish().describe(
12266
+ "Tells the speaker label model how many speakers it should attempt to identify. See [Set number of speakers expected](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization#set-number-of-speakers-expected) for more details."
12267
+ ),
12268
+ speech_models: import_zod3.z.array(
12269
+ import_zod3.z.string().describe(
12270
+ "The speech model to use for the transcription. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models."
12271
+ )
12272
+ ).describe(
12273
+ "List multiple speech models in priority order, allowing our system to automatically route your audio to the best available option. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models and routing behavior.\n"
12274
+ ),
12275
+ speech_threshold: import_zod3.z.number().nullish().describe(
12276
+ "Reject audio files that contain less than this fraction of speech.\nValid values are in the range [0, 1] inclusive. See [Speech Threshold](https://www.assemblyai.com/docs/speech-threshold) for more details.\n"
12277
+ ),
12278
+ speech_understanding: import_zod3.z.object({
12279
+ request: import_zod3.z.object({
12280
+ translation: import_zod3.z.object({
12281
+ target_languages: import_zod3.z.array(import_zod3.z.string()).describe(
12282
+ 'List of target language codes (e.g., `["es", "de"]`). See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for supported languages.'
12283
+ ),
12284
+ formal: import_zod3.z.boolean().default(createTranscriptBodySpeechUnderstandingRequestTranslationFormalDefault).describe(
12285
+ "Use formal language style. See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for more details."
12286
+ ),
12287
+ match_original_utterance: import_zod3.z.boolean().optional().describe(
12288
+ "When enabled with Speaker Labels, returns translated text in the utterances array. Each utterance will include a `translated_texts` key containing translations for each target language."
12289
+ )
12290
+ })
12291
+ }).describe(
12292
+ "Request body for [Translation](https://www.assemblyai.com/docs/speech-understanding/translation)."
12293
+ ).or(
12294
+ import_zod3.z.object({
12295
+ speaker_identification: import_zod3.z.object({
12296
+ speaker_type: import_zod3.z.enum(["role", "name"]).describe(
12297
+ "Type of speaker identification. See [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification) for details on each type."
12298
+ ),
12299
+ known_values: import_zod3.z.array(import_zod3.z.string()).optional().describe(
12300
+ 'Required if speaker_type is "role". Each value must be 35 characters or less.'
12301
+ ),
12302
+ speakers: import_zod3.z.array(
12303
+ import_zod3.z.object({
12304
+ role: import_zod3.z.string().optional().describe(
12305
+ 'The role of the speaker. Required when `speaker_type` is "role".'
12306
+ ),
12307
+ name: import_zod3.z.string().optional().describe(
12308
+ 'The name of the speaker. Required when `speaker_type` is "name".'
12309
+ ),
12310
+ description: import_zod3.z.string().optional().describe(
12311
+ "A description of the speaker to help the model identify them based on conversational context."
12312
+ )
12313
+ })
12314
+ ).optional().describe(
12315
+ "An array of speaker objects with metadata to improve identification accuracy. Each object should include a `role` or `name` (depending on `speaker_type`) and an optional `description` to help the model identify the speaker. You can also include any additional custom properties (e.g., `company`, `title`) to provide more context. Use this as an alternative to `known_values` when you want to provide additional context about each speaker."
12316
+ )
12317
+ })
12318
+ }).describe(
12319
+ "Request body for [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification)."
12320
+ )
12321
+ ).or(
12322
+ import_zod3.z.object({
12323
+ custom_formatting: import_zod3.z.object({
12324
+ date: import_zod3.z.string().optional().describe(
12325
+ 'Date format pattern (e.g., `"mm/dd/yyyy"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
12326
+ ),
12327
+ phone_number: import_zod3.z.string().optional().describe(
12328
+ 'Phone number format pattern (e.g., `"(xxx)xxx-xxxx"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
12329
+ ),
12330
+ email: import_zod3.z.string().optional().describe(
12331
+ 'Email format pattern (e.g., `"username@domain.com"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
12332
+ )
12333
+ })
12334
+ }).describe(
12335
+ "Request body for [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting)."
12336
+ )
12337
+ )
12338
+ }).optional().describe(
12339
+ "Enable speech understanding tasks like [Translation](https://www.assemblyai.com/docs/speech-understanding/translation), [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification), and [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting). See the task-specific docs for available options and configuration.\n"
12340
+ ),
12341
+ summarization: import_zod3.z.boolean().optional().describe(
12342
+ "Enable [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization), can be true or false. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details.\n\nNote: This parameter is only supported for the Universal-2 model.\n"
12343
+ ),
12344
+ summary_model: import_zod3.z.enum(["informative", "conversational", "catchy"]).optional().describe("The model to summarize the transcript"),
12345
+ summary_type: import_zod3.z.enum(["bullets", "bullets_verbose", "gist", "headline", "paragraph"]).optional().describe("The type of summary"),
12346
+ remove_audio_tags: import_zod3.z.enum(["all"]).describe(
12347
+ 'Remove [audio event tags](https://www.assemblyai.com/docs/pre-recorded-audio/universal-3-pro#audio-event-tags) from the transcript text. Set to `"all"` to remove all audio tags.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n'
12348
+ ).or(import_zod3.z.null()).optional().describe(
12349
+ 'Remove [audio event tags](https://www.assemblyai.com/docs/pre-recorded-audio/universal-3-pro#audio-event-tags) from the transcript text. Set to `"all"` to remove all audio tags.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n'
12350
+ ),
12351
+ temperature: import_zod3.z.number().optional().describe(
12352
+ "Control the amount of randomness injected into the model's response. See the [Prompting Guide](https://www.assemblyai.com/docs/pre-recorded-audio/prompting) for more details.\n\nNote: This parameter can only be used with the Universal-3 Pro model.\n"
12353
+ ),
12354
+ webhook_auth_header_name: import_zod3.z.string().nullish().describe(
12355
+ "The header name to be sent with the transcript completed or failed [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) requests"
12356
+ ),
12357
+ webhook_auth_header_value: import_zod3.z.string().nullish().describe(
12358
+ "The header value to send back with the transcript completed or failed [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) requests for added security"
12359
+ ),
12360
+ webhook_url: import_zod3.z.string().optional().describe(
12361
+ "The URL to which we send [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) requests.\n"
12362
+ ),
12363
+ custom_topics: import_zod3.z.boolean().optional().describe("This parameter does not currently have any functionality attached to it."),
12364
+ speech_model: import_zod3.z.string().describe(
12365
+ "The speech model to use for the transcription. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models."
12366
+ ).or(import_zod3.z.null()).optional().describe(
12367
+ "This parameter has been replaced with the `speech_models` parameter, learn more about the `speech_models` parameter [here](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model).\n"
12368
+ ),
12369
+ topics: import_zod3.z.array(import_zod3.z.string()).optional().describe("This parameter does not currently have any functionality attached to it."),
12370
+ audio_url: import_zod3.z.string().describe("The URL of the audio or video file to transcribe.")
12371
+ }).describe("The parameters for creating a transcript");
12372
+ var createTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault = "auto";
12373
+ var createTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault = false;
12374
+ var createTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault = 0.3;
12375
+ var createTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault = false;
12376
+ var createTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault = true;
12377
+ var createTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault = false;
12378
+ var createTranscriptResponse = import_zod3.z.object({
12379
+ audio_channels: import_zod3.z.number().optional().describe(
12380
+ "The number of audio channels in the audio file. This is only present when [multichannel](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) is enabled."
12381
+ ),
12382
+ audio_duration: import_zod3.z.number().nullish().describe("The duration of this transcript object's media file, in seconds"),
12383
+ audio_end_at: import_zod3.z.number().nullish().describe(
12384
+ "The point in time, in milliseconds, in the file at which the transcription was terminated. See [Set the start and end of the transcript](https://www.assemblyai.com/docs/pre-recorded-audio/set-the-start-and-end-of-the-transcript) for more details."
12385
+ ),
12386
+ audio_start_from: import_zod3.z.number().nullish().describe(
12387
+ "The point in time, in milliseconds, in the file at which the transcription was started. See [Set the start and end of the transcript](https://www.assemblyai.com/docs/pre-recorded-audio/set-the-start-and-end-of-the-transcript) for more details."
12388
+ ),
12389
+ audio_url: import_zod3.z.string().describe("The URL of the media that was transcribed"),
12390
+ auto_chapters: import_zod3.z.boolean().nullish().describe(
12391
+ "Whether [Auto Chapters](https://www.assemblyai.com/docs/speech-understanding/auto-chapters) is enabled, can be true or false. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible chapter summaries. See the [updated Auto Chapters page](https://www.assemblyai.com/docs/speech-understanding/auto-chapters) for details.\n\nNote: This parameter is only supported for the Universal-2 model.\n"
12392
+ ),
12393
+ auto_highlights: import_zod3.z.boolean().describe(
12394
+ "Whether [Key Phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases) is enabled, either true or false"
12395
+ ),
12396
+ auto_highlights_result: import_zod3.z.object({
12397
+ status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
12398
+ results: import_zod3.z.array(
12399
+ import_zod3.z.object({
12400
+ count: import_zod3.z.number().describe("The total number of times the key phrase appears in the audio file"),
12401
+ rank: import_zod3.z.number().describe(
12402
+ "The total relevancy to the overall audio file of this key phrase - a greater number means more relevant"
12403
+ ),
12404
+ text: import_zod3.z.string().describe("The text itself of the key phrase"),
12405
+ timestamps: import_zod3.z.array(
12406
+ import_zod3.z.object({
12407
+ start: import_zod3.z.number().describe("The start time in milliseconds"),
12408
+ end: import_zod3.z.number().describe("The end time in milliseconds")
12409
+ }).describe("Timestamp containing a start and end property in milliseconds")
12410
+ ).describe("The timestamp of the of the key phrase")
12411
+ })
12412
+ ).describe("A temporally-sequential array of Key Phrases")
12413
+ }).describe(
12414
+ "An array of results for the Key Phrases model, if it is enabled.\nSee [Key phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases) for more information.\n"
12415
+ ).or(import_zod3.z.null()).optional().describe(
12416
+ "An array of results for the Key Phrases model, if it is enabled.\nSee [Key Phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases) for more information.\n"
12417
+ ),
12418
+ chapters: import_zod3.z.array(
12419
+ import_zod3.z.object({
12420
+ gist: import_zod3.z.string().describe(
12421
+ "An ultra-short summary (just a few words) of the content spoken in the chapter"
12422
+ ),
12423
+ headline: import_zod3.z.string().describe("A single sentence summary of the content spoken during the chapter"),
12424
+ summary: import_zod3.z.string().describe("A one paragraph summary of the content spoken during the chapter"),
12425
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter"),
12426
+ end: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter")
12427
+ }).describe("Chapter of the audio file")
12428
+ ).nullish().describe(
12429
+ "An array of temporally sequential chapters for the audio file. See [Auto Chapters](https://www.assemblyai.com/docs/speech-understanding/auto-chapters) for more information."
12430
+ ),
12431
+ confidence: import_zod3.z.number().nullish().describe(
12432
+ "The confidence score for the transcript, between 0.0 (low confidence) and 1.0 (high confidence)"
12433
+ ),
12434
+ content_safety: import_zod3.z.boolean().nullish().describe(
12435
+ "Whether [Content Moderation](https://www.assemblyai.com/docs/content-moderation) is enabled, can be true or false"
12436
+ ),
12437
+ content_safety_labels: import_zod3.z.object({
12438
+ status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
12439
+ results: import_zod3.z.array(
12440
+ import_zod3.z.object({
12441
+ text: import_zod3.z.string().describe("The transcript of the section flagged by the Content Moderation model"),
12442
+ labels: import_zod3.z.array(
12443
+ import_zod3.z.object({
12444
+ label: import_zod3.z.string().describe("The label of the sensitive topic"),
12445
+ confidence: import_zod3.z.number().describe("The confidence score for the topic being discussed, from 0 to 1"),
12446
+ severity: import_zod3.z.number().describe("How severely the topic is discussed in the section, from 0 to 1")
12447
+ })
12448
+ ).describe(
12449
+ "An array of safety labels, one per sensitive topic that was detected in the section"
12450
+ ),
12451
+ sentences_idx_start: import_zod3.z.number().describe("The sentence index at which the section begins"),
12452
+ sentences_idx_end: import_zod3.z.number().describe("The sentence index at which the section ends"),
12453
+ timestamp: import_zod3.z.object({
12454
+ start: import_zod3.z.number().describe("The start time in milliseconds"),
12455
+ end: import_zod3.z.number().describe("The end time in milliseconds")
12456
+ }).describe("Timestamp containing a start and end property in milliseconds")
12457
+ })
12458
+ ).describe("An array of results for the Content Moderation model"),
12459
+ summary: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.number()).describe(
12460
+ "A summary of the Content Moderation confidence results for the entire audio file"
12014
12461
  ),
12015
- redact_pii_policies: import_zod3.z.array(
12016
- import_zod3.z.enum([
12462
+ severity_score_summary: import_zod3.z.record(
12463
+ import_zod3.z.string(),
12464
+ import_zod3.z.object({
12465
+ low: import_zod3.z.number(),
12466
+ medium: import_zod3.z.number(),
12467
+ high: import_zod3.z.number()
12468
+ })
12469
+ ).describe(
12470
+ "A summary of the Content Moderation severity results for the entire audio file"
12471
+ )
12472
+ }).describe(
12473
+ "An array of results for the Content Moderation model, if it is enabled.\nSee [Content moderation](https://www.assemblyai.com/docs/content-moderation) for more information.\n"
12474
+ ).or(import_zod3.z.null()).optional().describe(
12475
+ "An array of results for the Content Moderation model, if it is enabled.\nSee [Content moderation](https://www.assemblyai.com/docs/content-moderation) for more information.\n"
12476
+ ),
12477
+ custom_spelling: import_zod3.z.array(
12478
+ import_zod3.z.object({
12479
+ from: import_zod3.z.array(import_zod3.z.string()).describe("Words or phrases to replace"),
12480
+ to: import_zod3.z.string().describe("Word to replace with")
12481
+ }).describe(
12482
+ "Object containing words or phrases to replace, and the word or phrase to replace with"
12483
+ )
12484
+ ).nullish().describe(
12485
+ "Customize how words are spelled and formatted using to and from values. See [Custom Spelling](https://www.assemblyai.com/docs/pre-recorded-audio/custom-spelling) for more details."
12486
+ ),
12487
+ disfluencies: import_zod3.z.boolean().nullish().describe(
12488
+ 'Transcribe [Filler Words](https://www.assemblyai.com/docs/pre-recorded-audio/filler-words), like "umm", in your media file; can be true or false'
12489
+ ),
12490
+ domain: import_zod3.z.string().nullish().describe(
12491
+ 'The domain-specific model applied to the transcript. When set to `"medical-v1"`, [Medical Mode](https://www.assemblyai.com/docs/pre-recorded-audio/medical-mode) was used to improve accuracy for medical terminology.\n'
12492
+ ),
12493
+ entities: import_zod3.z.array(
12494
+ import_zod3.z.object({
12495
+ entity_type: import_zod3.z.enum([
12017
12496
  "account_number",
12018
12497
  "banking_information",
12019
12498
  "blood_type",
@@ -12058,103 +12537,59 @@ var createTranscriptBody = import_zod3.z.object({
12058
12537
  "username",
12059
12538
  "vehicle_id",
12060
12539
  "zodiac_sign"
12061
- ]).describe("The type of PII to redact")
12062
- ).optional().describe(
12063
- "The list of PII Redaction policies to enable. See [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more details."
12064
- ),
12065
- redact_pii_sub: import_zod3.z.enum(["entity_name", "hash"]).describe(
12066
- 'The replacement logic for detected PII, can be "entity_name" or "hash". See [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more details.'
12067
- ).or(import_zod3.z.null()).default(createTranscriptBodyRedactPiiSubDefault).describe(
12068
- 'The replacement logic for detected PII, can be "entity_type" or "hash". See [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more details.'
12069
- ),
12070
- speaker_labels: import_zod3.z.boolean().optional().describe(
12071
- "Enable [Speaker diarization](https://www.assemblyai.com/docs/models/speaker-diarization), can be true or false"
12072
- ),
12073
- speakers_expected: import_zod3.z.number().nullish().describe(
12074
- "Tells the speaker label model how many speakers it should attempt to identify. See [Speaker diarization](https://www.assemblyai.com/docs/models/speaker-diarization) for more details."
12075
- ),
12076
- content_safety: import_zod3.z.boolean().optional().describe(
12077
- "Enable [Content Moderation](https://www.assemblyai.com/docs/models/content-moderation), can be true or false"
12078
- ),
12079
- content_safety_confidence: import_zod3.z.number().min(createTranscriptBodyContentSafetyConfidenceMin).max(createTranscriptBodyContentSafetyConfidenceMax).default(createTranscriptBodyContentSafetyConfidenceDefault).describe(
12080
- "The confidence threshold for the Content Moderation model. Values must be between 25 and 100."
12081
- ),
12082
- iab_categories: import_zod3.z.boolean().optional().describe(
12083
- "Enable [Topic Detection](https://www.assemblyai.com/docs/models/topic-detection), can be true or false"
12084
- ),
12085
- custom_spelling: import_zod3.z.array(
12086
- import_zod3.z.object({
12087
- from: import_zod3.z.array(import_zod3.z.string().describe("Word or phrase to replace")).describe("Words or phrases to replace"),
12088
- to: import_zod3.z.string().describe("Word to replace with")
12089
- }).describe(
12090
- "Object containing words or phrases to replace, and the word or phrase to replace with"
12540
+ ]).describe("The type of entity for the detected entity"),
12541
+ text: import_zod3.z.string().describe("The text for the detected entity"),
12542
+ start: import_zod3.z.number().describe(
12543
+ "The starting time, in milliseconds, at which the detected entity appears in the audio file"
12544
+ ),
12545
+ end: import_zod3.z.number().describe(
12546
+ "The ending time, in milliseconds, for the detected entity in the audio file"
12091
12547
  )
12092
- ).optional().describe("Customize how words are spelled and formatted using to and from values"),
12093
- keyterms_prompt: import_zod3.z.array(import_zod3.z.string()).optional().describe(
12094
- "<Warning>`keyterms_prompt` is only supported when the `speech_model` is specified as `slam-1`</Warning>\nImprove accuracy with up to 1000 domain-specific words or phrases (maximum 6 words per phrase).\n"
12095
- ),
12096
- prompt: import_zod3.z.string().optional().describe("This parameter does not currently have any functionality attached to it."),
12097
- sentiment_analysis: import_zod3.z.boolean().optional().describe(
12098
- "Enable [Sentiment Analysis](https://www.assemblyai.com/docs/models/sentiment-analysis), can be true or false"
12099
- ),
12100
- auto_chapters: import_zod3.z.boolean().optional().describe(
12101
- "Enable [Auto Chapters](https://www.assemblyai.com/docs/models/auto-chapters), can be true or false"
12102
- ),
12103
- entity_detection: import_zod3.z.boolean().optional().describe(
12104
- "Enable [Entity Detection](https://www.assemblyai.com/docs/models/entity-detection), can be true or false"
12105
- ),
12106
- speech_threshold: import_zod3.z.number().min(createTranscriptBodySpeechThresholdMin).max(createTranscriptBodySpeechThresholdMax).nullish().describe(
12107
- "Reject audio files that contain less than this fraction of speech.\nValid values are in the range [0, 1] inclusive.\n"
12108
- ),
12109
- summarization: import_zod3.z.boolean().optional().describe(
12110
- "Enable [Summarization](https://www.assemblyai.com/docs/models/summarization), can be true or false"
12111
- ),
12112
- summary_model: import_zod3.z.enum(["informative", "conversational", "catchy"]).optional().describe("The model to summarize the transcript"),
12113
- summary_type: import_zod3.z.enum(["bullets", "bullets_verbose", "gist", "headline", "paragraph"]).optional().describe("The type of summary"),
12114
- custom_topics: import_zod3.z.boolean().optional().describe("Enable custom topics, either true or false"),
12115
- topics: import_zod3.z.array(import_zod3.z.string()).optional().describe("The list of custom topics")
12116
- }).describe("The parameters for creating a transcript")
12117
- ).describe("The parameters for creating a transcript");
12118
- var createTranscriptResponseLanguageConfidenceThresholdMin = 0;
12119
- var createTranscriptResponseLanguageConfidenceThresholdMax = 1;
12120
- var createTranscriptResponseLanguageConfidenceMin = 0;
12121
- var createTranscriptResponseLanguageConfidenceMax = 1;
12122
- var createTranscriptResponseSpeechModelDefault = null;
12123
- var createTranscriptResponseWordsItemConfidenceMin = 0;
12124
- var createTranscriptResponseWordsItemConfidenceMax = 1;
12125
- var createTranscriptResponseUtterancesItemConfidenceMin = 0;
12126
- var createTranscriptResponseUtterancesItemConfidenceMax = 1;
12127
- var createTranscriptResponseUtterancesItemWordsItemConfidenceMin = 0;
12128
- var createTranscriptResponseUtterancesItemWordsItemConfidenceMax = 1;
12129
- var createTranscriptResponseConfidenceMin = 0;
12130
- var createTranscriptResponseConfidenceMax = 1;
12131
- var createTranscriptResponseAutoHighlightsResultResultsItemRankMin = 0;
12132
- var createTranscriptResponseAutoHighlightsResultResultsItemRankMax = 1;
12133
- var createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin = 0;
12134
- var createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax = 1;
12135
- var createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin = 0;
12136
- var createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax = 1;
12137
- var createTranscriptResponseContentSafetyLabelsSummaryMinOne = 0;
12138
- var createTranscriptResponseContentSafetyLabelsSummaryMaxOne = 1;
12139
- var createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin = 0;
12140
- var createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax = 1;
12141
- var createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin = 0;
12142
- var createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax = 1;
12143
- var createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin = 0;
12144
- var createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax = 1;
12145
- var createTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin = 0;
12146
- var createTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax = 1;
12147
- var createTranscriptResponseIabCategoriesResultSummaryMinOne = 0;
12148
- var createTranscriptResponseIabCategoriesResultSummaryMaxOne = 1;
12149
- var createTranscriptResponseSentimentAnalysisResultsItemConfidenceMin = 0;
12150
- var createTranscriptResponseSentimentAnalysisResultsItemConfidenceMax = 1;
12151
- var createTranscriptResponseSpeechThresholdMin = 0;
12152
- var createTranscriptResponseSpeechThresholdMax = 1;
12153
- var createTranscriptResponse = import_zod3.z.object({
12548
+ }).describe("A detected entity")
12549
+ ).nullish().describe(
12550
+ "An array of results for the Entity Detection model, if it is enabled.\nSee [Entity detection](https://www.assemblyai.com/docs/speech-understanding/entity-detection) for more information.\n"
12551
+ ),
12552
+ entity_detection: import_zod3.z.boolean().nullish().describe(
12553
+ "Whether [Entity Detection](https://www.assemblyai.com/docs/speech-understanding/entity-detection) is enabled, can be true or false"
12554
+ ),
12555
+ error: import_zod3.z.string().optional().describe("Error message of why the transcript failed"),
12556
+ filter_profanity: import_zod3.z.boolean().nullish().describe(
12557
+ "Whether [Profanity Filtering](https://www.assemblyai.com/docs/profanity-filtering) is enabled, either true or false"
12558
+ ),
12559
+ format_text: import_zod3.z.boolean().nullish().describe(
12560
+ "Whether [Text Formatting](https://www.assemblyai.com/docs/pre-recorded-audio) is enabled, either true or false"
12561
+ ),
12562
+ iab_categories: import_zod3.z.boolean().nullish().describe(
12563
+ "Whether [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection) is enabled, can be true or false"
12564
+ ),
12565
+ iab_categories_result: import_zod3.z.object({
12566
+ status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
12567
+ results: import_zod3.z.array(
12568
+ import_zod3.z.object({
12569
+ text: import_zod3.z.string().describe("The text in the transcript in which a detected topic occurs"),
12570
+ labels: import_zod3.z.array(
12571
+ import_zod3.z.object({
12572
+ relevance: import_zod3.z.number().describe("How relevant the detected topic is of a detected topic"),
12573
+ label: import_zod3.z.string().describe(
12574
+ "The IAB taxonomical label for the label of the detected topic, where > denotes supertopic/subtopic relationship"
12575
+ )
12576
+ })
12577
+ ).optional().describe("An array of detected topics in the text"),
12578
+ timestamp: import_zod3.z.object({
12579
+ start: import_zod3.z.number().describe("The start time in milliseconds"),
12580
+ end: import_zod3.z.number().describe("The end time in milliseconds")
12581
+ }).optional().describe("Timestamp containing a start and end property in milliseconds")
12582
+ }).describe("The result of the topic detection model")
12583
+ ).describe("An array of results for the Topic Detection model"),
12584
+ summary: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.number()).describe("The overall relevance of topic to the entire audio file")
12585
+ }).describe(
12586
+ "The result of the Topic Detection model, if it is enabled.\nSee [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection) for more information.\n"
12587
+ ).or(import_zod3.z.null()).optional().describe(
12588
+ "The result of the Topic Detection model, if it is enabled.\nSee [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection) for more information.\n"
12589
+ ),
12154
12590
  id: import_zod3.z.string().uuid().describe("The unique identifier of your transcript"),
12155
- audio_url: import_zod3.z.string().describe("The URL of the media that was transcribed"),
12156
- status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).describe(
12157
- "The status of your transcript. Possible values are queued, processing, completed, or error."
12591
+ keyterms_prompt: import_zod3.z.array(import_zod3.z.string()).optional().describe(
12592
+ "Improve accuracy with up to 200 (for Universal-2) or 1000 (for Universal-3 Pro) domain-specific words or phrases (maximum 6 words per phrase). See [Keyterms Prompting](https://www.assemblyai.com/docs/pre-recorded-audio/keyterms-prompting) for more details.\n"
12158
12593
  ),
12159
12594
  language_code: import_zod3.z.enum([
12160
12595
  "en",
@@ -12259,138 +12694,175 @@ var createTranscriptResponse = import_zod3.z.object({
12259
12694
  "cy",
12260
12695
  "yi",
12261
12696
  "yo"
12262
- ]).describe(
12263
- "The language of your audio file. Possible values are found in [Supported Languages](https://www.assemblyai.com/docs/concepts/supported-languages).\nThe default value is 'en_us'.\n"
12264
- ).or(import_zod3.z.string()).optional().describe(
12265
- "The language of your audio file.\nPossible values are found in [Supported Languages](https://www.assemblyai.com/docs/concepts/supported-languages).\nThe default value is 'en_us'.\n"
12266
- ),
12267
- language_detection: import_zod3.z.boolean().nullish().describe(
12268
- "Whether [Automatic language detection](https://www.assemblyai.com/docs/models/speech-recognition#automatic-language-detection) is enabled, either true or false"
12269
- ),
12270
- language_confidence_threshold: import_zod3.z.number().min(createTranscriptResponseLanguageConfidenceThresholdMin).max(createTranscriptResponseLanguageConfidenceThresholdMax).nullable().describe(
12271
- "The confidence threshold for the automatically detected language.\nAn error will be returned if the language confidence is below this threshold.\n"
12272
- ),
12273
- language_confidence: import_zod3.z.number().min(createTranscriptResponseLanguageConfidenceMin).max(createTranscriptResponseLanguageConfidenceMax).nullable().describe(
12274
- "The confidence score for the detected language, between 0.0 (low confidence) and 1.0 (high confidence)"
12275
- ),
12276
- speech_model: import_zod3.z.enum(["best", "slam-1", "universal"]).describe("The speech model to use for the transcription.").or(import_zod3.z.null()).describe(
12277
- "The speech model used for the transcription. When `null`, the default model is used."
12278
- ),
12279
- text: import_zod3.z.string().nullish().describe("The textual transcript of your media file"),
12280
- words: import_zod3.z.array(
12281
- import_zod3.z.object({
12282
- confidence: import_zod3.z.number().min(createTranscriptResponseWordsItemConfidenceMin).max(createTranscriptResponseWordsItemConfidenceMax).describe("The confidence score for the transcript of this word"),
12283
- start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
12284
- end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
12285
- text: import_zod3.z.string().describe("The text of the word"),
12286
- channel: import_zod3.z.string().nullish().describe(
12287
- "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
12288
- ),
12289
- speaker: import_zod3.z.string().nullable().describe(
12290
- "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
12291
- )
12292
- })
12293
- ).nullish().describe(
12294
- "An array of temporally-sequential word objects, one for each word in the transcript.\nSee [Speech recognition](https://www.assemblyai.com/docs/models/speech-recognition) for more information.\n"
12697
+ ]).optional().describe(
12698
+ "The language of your audio file. Possible values are found in [Supported Languages](https://www.assemblyai.com/docs/pre-recorded-audio/supported-languages).\nThe default value is 'en_us'.\n"
12295
12699
  ),
12296
- utterances: import_zod3.z.array(
12297
- import_zod3.z.object({
12298
- confidence: import_zod3.z.number().min(createTranscriptResponseUtterancesItemConfidenceMin).max(createTranscriptResponseUtterancesItemConfidenceMax).describe("The confidence score for the transcript of this utterance"),
12299
- start: import_zod3.z.number().describe("The starting time, in milliseconds, of the utterance in the audio file"),
12300
- end: import_zod3.z.number().describe("The ending time, in milliseconds, of the utterance in the audio file"),
12301
- text: import_zod3.z.string().describe("The text for this utterance"),
12302
- words: import_zod3.z.array(
12303
- import_zod3.z.object({
12304
- confidence: import_zod3.z.number().min(createTranscriptResponseUtterancesItemWordsItemConfidenceMin).max(createTranscriptResponseUtterancesItemWordsItemConfidenceMax).describe("The confidence score for the transcript of this word"),
12305
- start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
12306
- end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
12307
- text: import_zod3.z.string().describe("The text of the word"),
12308
- channel: import_zod3.z.string().nullish().describe(
12309
- "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
12310
- ),
12311
- speaker: import_zod3.z.string().nullable().describe(
12312
- "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
12313
- )
12314
- })
12315
- ).describe("The words in the utterance."),
12316
- channel: import_zod3.z.string().nullish().describe(
12317
- "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
12318
- ),
12319
- speaker: import_zod3.z.string().describe(
12320
- 'The speaker of this utterance, where each speaker is assigned a sequential capital letter - e.g. "A" for Speaker A, "B" for Speaker B, etc.'
12321
- )
12322
- })
12700
+ language_codes: import_zod3.z.array(
12701
+ import_zod3.z.enum([
12702
+ "en",
12703
+ "en_au",
12704
+ "en_uk",
12705
+ "en_us",
12706
+ "es",
12707
+ "fr",
12708
+ "de",
12709
+ "it",
12710
+ "pt",
12711
+ "nl",
12712
+ "af",
12713
+ "sq",
12714
+ "am",
12715
+ "ar",
12716
+ "hy",
12717
+ "as",
12718
+ "az",
12719
+ "ba",
12720
+ "eu",
12721
+ "be",
12722
+ "bn",
12723
+ "bs",
12724
+ "br",
12725
+ "bg",
12726
+ "my",
12727
+ "ca",
12728
+ "zh",
12729
+ "hr",
12730
+ "cs",
12731
+ "da",
12732
+ "et",
12733
+ "fo",
12734
+ "fi",
12735
+ "gl",
12736
+ "ka",
12737
+ "el",
12738
+ "gu",
12739
+ "ht",
12740
+ "ha",
12741
+ "haw",
12742
+ "he",
12743
+ "hi",
12744
+ "hu",
12745
+ "is",
12746
+ "id",
12747
+ "ja",
12748
+ "jw",
12749
+ "kn",
12750
+ "kk",
12751
+ "km",
12752
+ "ko",
12753
+ "lo",
12754
+ "la",
12755
+ "lv",
12756
+ "ln",
12757
+ "lt",
12758
+ "lb",
12759
+ "mk",
12760
+ "mg",
12761
+ "ms",
12762
+ "ml",
12763
+ "mt",
12764
+ "mi",
12765
+ "mr",
12766
+ "mn",
12767
+ "ne",
12768
+ "no",
12769
+ "nn",
12770
+ "oc",
12771
+ "pa",
12772
+ "ps",
12773
+ "fa",
12774
+ "pl",
12775
+ "ro",
12776
+ "ru",
12777
+ "sa",
12778
+ "sr",
12779
+ "sn",
12780
+ "sd",
12781
+ "si",
12782
+ "sk",
12783
+ "sl",
12784
+ "so",
12785
+ "su",
12786
+ "sw",
12787
+ "sv",
12788
+ "tl",
12789
+ "tg",
12790
+ "ta",
12791
+ "tt",
12792
+ "te",
12793
+ "th",
12794
+ "bo",
12795
+ "tr",
12796
+ "tk",
12797
+ "uk",
12798
+ "ur",
12799
+ "uz",
12800
+ "vi",
12801
+ "cy",
12802
+ "yi",
12803
+ "yo"
12804
+ ]).describe(
12805
+ "The language of your audio file. Possible values are found in [Supported Languages](https://www.assemblyai.com/docs/pre-recorded-audio/supported-languages).\nThe default value is 'en_us'.\n"
12806
+ )
12323
12807
  ).nullish().describe(
12324
- "When multichannel or speaker_labels is enabled, a list of turn-by-turn utterance objects.\nSee [Speaker diarization](https://www.assemblyai.com/docs/speech-to-text/speaker-diarization) and [Multichannel transcription](https://www.assemblyai.com/docs/speech-to-text/speech-recognition#multichannel-transcription) for more information.\n"
12808
+ "The language codes of your audio file. Used for [Code switching](/docs/speech-to-text/pre-recorded-audio/code-switching)\nOne of the values specified must be `en`.\n"
12325
12809
  ),
12326
- confidence: import_zod3.z.number().min(createTranscriptResponseConfidenceMin).max(createTranscriptResponseConfidenceMax).nullish().describe(
12327
- "The confidence score for the transcript, between 0.0 (low confidence) and 1.0 (high confidence)"
12810
+ language_confidence: import_zod3.z.number().nullable().describe(
12811
+ "The confidence score for the detected language, between 0.0 (low confidence) and 1.0 (high confidence). See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details."
12328
12812
  ),
12329
- audio_duration: import_zod3.z.number().nullish().describe("The duration of this transcript object's media file, in seconds"),
12330
- punctuate: import_zod3.z.boolean().nullish().describe("Whether Automatic Punctuation is enabled, either true or false"),
12331
- format_text: import_zod3.z.boolean().nullish().describe("Whether Text Formatting is enabled, either true or false"),
12332
- disfluencies: import_zod3.z.boolean().nullish().describe('Transcribe Filler Words, like "umm", in your media file; can be true or false'),
12333
- multichannel: import_zod3.z.boolean().nullish().describe(
12334
- "Whether [Multichannel transcription](https://www.assemblyai.com/docs/models/speech-recognition#multichannel-transcription) was enabled in the transcription request, either true or false"
12813
+ language_confidence_threshold: import_zod3.z.number().nullable().describe(
12814
+ "The confidence threshold for the automatically detected language.\nAn error will be returned if the language confidence is below this threshold.\nSee [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.\n"
12335
12815
  ),
12336
- audio_channels: import_zod3.z.number().optional().describe(
12337
- "The number of audio channels in the audio file. This is only present when multichannel is enabled."
12338
- ),
12339
- webhook_url: import_zod3.z.string().nullish().describe(
12340
- "The URL to which we send webhook requests.\nWe sends two different types of webhook requests.\nOne request when a transcript is completed or failed, and one request when the redacted audio is ready if redact_pii_audio is enabled.\n"
12341
- ),
12342
- webhook_status_code: import_zod3.z.number().nullish().describe(
12343
- "The status code we received from your server when delivering the transcript completed or failed webhook request, if a webhook URL was provided"
12816
+ language_detection: import_zod3.z.boolean().nullish().describe(
12817
+ "Whether [Automatic language detection](/docs/pre-recorded-audio/automatic-language-detection) is enabled, either true or false"
12344
12818
  ),
12345
- webhook_auth: import_zod3.z.boolean().describe("Whether webhook authentication details were provided"),
12346
- webhook_auth_header_name: import_zod3.z.string().nullish().describe(
12347
- "The header name to be sent with the transcript completed or failed webhook requests"
12819
+ language_detection_options: import_zod3.z.object({
12820
+ expected_languages: import_zod3.z.array(import_zod3.z.string()).optional().describe(
12821
+ 'List of languages expected in the audio file. Defaults to `["all"]` when unspecified. See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.'
12822
+ ),
12823
+ fallback_language: import_zod3.z.string().default(createTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault).describe(
12824
+ 'If the detected language of the audio file is not in the list of expected languages, the `fallback_language` is used. Specify `["auto"]` to let our model choose the fallback language from `expected_languages` with the highest confidence score. See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.\n'
12825
+ ),
12826
+ code_switching: import_zod3.z.boolean().optional().describe(
12827
+ "Whether [code switching](/docs/speech-to-text/pre-recorded-audio/code-switching) should be detected.\n"
12828
+ ),
12829
+ code_switching_confidence_threshold: import_zod3.z.number().default(
12830
+ createTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault
12831
+ ).describe(
12832
+ "The confidence threshold for [code switching](/docs/speech-to-text/pre-recorded-audio/code-switching) detection. If the code switching confidence is below this threshold, the transcript will be processed in the language with the highest `language_detection_confidence` score.\n"
12833
+ )
12834
+ }).optional().describe(
12835
+ "Specify options for [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection)."
12348
12836
  ),
12349
- speed_boost: import_zod3.z.boolean().nullish().describe("Whether speed boost is enabled"),
12350
- auto_highlights: import_zod3.z.boolean().describe("Whether Key Phrases is enabled, either true or false"),
12351
- auto_highlights_result: import_zod3.z.object({
12352
- status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
12353
- results: import_zod3.z.array(
12354
- import_zod3.z.object({
12355
- count: import_zod3.z.number().describe("The total number of times the key phrase appears in the audio file"),
12356
- rank: import_zod3.z.number().min(createTranscriptResponseAutoHighlightsResultResultsItemRankMin).max(createTranscriptResponseAutoHighlightsResultResultsItemRankMax).describe(
12357
- "The total relevancy to the overall audio file of this key phrase - a greater number means more relevant"
12358
- ),
12359
- text: import_zod3.z.string().describe("The text itself of the key phrase"),
12360
- timestamps: import_zod3.z.array(
12361
- import_zod3.z.object({
12362
- start: import_zod3.z.number().describe("The start time in milliseconds"),
12363
- end: import_zod3.z.number().describe("The end time in milliseconds")
12364
- }).describe("Timestamp containing a start and end property in milliseconds")
12365
- ).describe("The timestamp of the of the key phrase")
12366
- })
12367
- ).describe("A temporally-sequential array of Key Phrases")
12368
- }).describe(
12369
- "An array of results for the Key Phrases model, if it is enabled.\nSee [Key phrases](https://www.assemblyai.com/docs/models/key-phrases) for more information.\n"
12370
- ).or(import_zod3.z.null()).optional().describe(
12371
- "An array of results for the Key Phrases model, if it is enabled.\nSee [Key Phrases](https://www.assemblyai.com/docs/models/key-phrases) for more information.\n"
12837
+ multichannel: import_zod3.z.boolean().nullish().describe(
12838
+ "Whether [Multichannel transcription](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) was enabled in the transcription request, either true or false"
12372
12839
  ),
12373
- audio_start_from: import_zod3.z.number().nullish().describe(
12374
- "The point in time, in milliseconds, in the file at which the transcription was started"
12840
+ prompt: import_zod3.z.string().optional().describe(
12841
+ "Provide natural language prompting of up to 1,500 words of contextual information to the model. See the [Prompting Guide](https://www.assemblyai.com/docs/pre-recorded-audio/prompting) for best practices.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n"
12375
12842
  ),
12376
- audio_end_at: import_zod3.z.number().nullish().describe(
12377
- "The point in time, in milliseconds, in the file at which the transcription was terminated"
12378
- ),
12379
- word_boost: import_zod3.z.array(import_zod3.z.string()).optional().describe("The list of custom vocabulary to boost transcription probability for"),
12380
- boost_param: import_zod3.z.string().nullish().describe("The word boost parameter value"),
12381
- filter_profanity: import_zod3.z.boolean().nullish().describe(
12382
- "Whether [Profanity Filtering](https://www.assemblyai.com/docs/models/speech-recognition#profanity-filtering) is enabled, either true or false"
12843
+ punctuate: import_zod3.z.boolean().nullish().describe(
12844
+ "Whether [Automatic Punctuation](https://www.assemblyai.com/docs/pre-recorded-audio) is enabled, either true or false"
12383
12845
  ),
12384
12846
  redact_pii: import_zod3.z.boolean().describe(
12385
- "Whether [PII Redaction](https://www.assemblyai.com/docs/models/pii-redaction) is enabled, either true or false"
12847
+ "Whether [PII Redaction](https://www.assemblyai.com/docs/pii-redaction) is enabled, either true or false"
12386
12848
  ),
12387
12849
  redact_pii_audio: import_zod3.z.boolean().nullish().describe(
12388
- "Whether a redacted version of the audio file was generated,\neither true or false. See [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more information.\n"
12850
+ "Whether a redacted version of the audio file was generated,\neither true or false. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more information.\n"
12851
+ ),
12852
+ redact_pii_audio_options: import_zod3.z.object({
12853
+ return_redacted_no_speech_audio: import_zod3.z.boolean().optional().describe(
12854
+ "By default, audio redaction provides redacted audio URLs only when speech is detected. However, if your use-case specifically requires redacted audio files even for silent audio files without any dialogue, you can opt to receive these URLs by setting this parameter to `true`."
12855
+ ),
12856
+ override_audio_redaction_method: import_zod3.z.enum(["silence"]).optional().describe(
12857
+ "Specify the method used to redact audio. By default, redacted audio uses a beep sound. Set to `silence` to replace PII with silence instead of a beep."
12858
+ )
12859
+ }).optional().describe(
12860
+ "The options for PII-redacted audio, if redact_pii_audio is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more information.\n"
12389
12861
  ),
12390
12862
  redact_pii_audio_quality: import_zod3.z.enum(["mp3", "wav"]).describe(
12391
- "Controls the filetype of the audio created by redact_pii_audio. Currently supports mp3 (default) and wav. See [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more details."
12863
+ "Controls the filetype of the audio created by redact_pii_audio. Currently supports mp3 (default) and wav. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more details."
12392
12864
  ).or(import_zod3.z.null()).optional().describe(
12393
- "The audio quality of the PII-redacted audio file, if redact_pii_audio is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more information.\n"
12865
+ "The audio quality of the PII-redacted audio file, if redact_pii_audio is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more information.\n"
12394
12866
  ),
12395
12867
  redact_pii_policies: import_zod3.z.array(
12396
12868
  import_zod3.z.enum([
@@ -12440,19 +12912,347 @@ var createTranscriptResponse = import_zod3.z.object({
12440
12912
  "zodiac_sign"
12441
12913
  ]).describe("The type of PII to redact")
12442
12914
  ).nullish().describe(
12443
- "The list of PII Redaction policies that were enabled, if PII Redaction is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more information.\n"
12915
+ "The list of PII Redaction policies that were enabled, if PII Redaction is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/pii-redaction) for more information.\n"
12444
12916
  ),
12445
12917
  redact_pii_sub: import_zod3.z.enum(["entity_name", "hash"]).optional().describe(
12446
- 'The replacement logic for detected PII, can be "entity_name" or "hash". See [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more details.'
12918
+ "The replacement logic for detected PII, can be `entity_name` or `hash`. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction) for more details."
12919
+ ),
12920
+ sentiment_analysis: import_zod3.z.boolean().nullish().describe(
12921
+ "Whether [Sentiment Analysis](https://www.assemblyai.com/docs/speech-understanding/sentiment-analysis) is enabled, can be true or false"
12922
+ ),
12923
+ sentiment_analysis_results: import_zod3.z.array(
12924
+ import_zod3.z.object({
12925
+ text: import_zod3.z.string().describe("The transcript of the sentence"),
12926
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, of the sentence"),
12927
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, of the sentence"),
12928
+ sentiment: import_zod3.z.enum(["POSITIVE", "NEUTRAL", "NEGATIVE"]),
12929
+ confidence: import_zod3.z.number().describe(
12930
+ "The confidence score for the detected sentiment of the sentence, from 0 to 1"
12931
+ ),
12932
+ channel: import_zod3.z.string().nullish().describe(
12933
+ "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
12934
+ ),
12935
+ speaker: import_zod3.z.string().nullable().describe(
12936
+ "The speaker of the sentence if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
12937
+ )
12938
+ }).describe("The result of the Sentiment Analysis model")
12939
+ ).nullish().describe(
12940
+ "An array of results for the Sentiment Analysis model, if it is enabled.\nSee [Sentiment Analysis](https://www.assemblyai.com/docs/speech-understanding/sentiment-analysis) for more information.\n"
12447
12941
  ),
12448
12942
  speaker_labels: import_zod3.z.boolean().nullish().describe(
12449
- "Whether [Speaker diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, can be true or false"
12943
+ "Whether [Speaker diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, can be true or false"
12450
12944
  ),
12451
12945
  speakers_expected: import_zod3.z.number().nullish().describe(
12452
- "Tell the speaker label model how many speakers it should attempt to identify. See [Speaker diarization](https://www.assemblyai.com/docs/models/speaker-diarization) for more details."
12946
+ "Tell the speaker label model how many speakers it should attempt to identify. See [Set number of speakers expected](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization#set-number-of-speakers-expected) for more details."
12947
+ ),
12948
+ speech_model_used: import_zod3.z.string().optional().describe(
12949
+ "The speech model to use for the transcription. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models."
12950
+ ),
12951
+ speech_models: import_zod3.z.array(
12952
+ import_zod3.z.string().describe(
12953
+ "The speech model to use for the transcription. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models."
12954
+ )
12955
+ ).nullish().describe(
12956
+ "List multiple speech models in priority order, allowing our system to automatically route your audio to the best available option. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models and routing behavior.\n"
12957
+ ),
12958
+ speech_threshold: import_zod3.z.number().nullish().describe(
12959
+ "Defaults to null. Reject audio files that contain less than this fraction of speech.\nValid values are in the range [0, 1] inclusive. See [Speech Threshold](https://www.assemblyai.com/docs/speech-threshold) for more details.\n"
12960
+ ),
12961
+ speech_understanding: import_zod3.z.object({
12962
+ request: import_zod3.z.object({
12963
+ translation: import_zod3.z.object({
12964
+ target_languages: import_zod3.z.array(import_zod3.z.string()).describe(
12965
+ 'List of target language codes (e.g., `["es", "de"]`). See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for supported languages.'
12966
+ ),
12967
+ formal: import_zod3.z.boolean().default(createTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault).describe(
12968
+ "Use formal language style. See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for more details."
12969
+ ),
12970
+ match_original_utterance: import_zod3.z.boolean().optional().describe(
12971
+ "When enabled with Speaker Labels, returns translated text in the utterances array. Each utterance will include a `translated_texts` key containing translations for each target language."
12972
+ )
12973
+ })
12974
+ }).describe(
12975
+ "Request body for [Translation](https://www.assemblyai.com/docs/speech-understanding/translation)."
12976
+ ).or(
12977
+ import_zod3.z.object({
12978
+ speaker_identification: import_zod3.z.object({
12979
+ speaker_type: import_zod3.z.enum(["role", "name"]).describe(
12980
+ "Type of speaker identification. See [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification) for details on each type."
12981
+ ),
12982
+ known_values: import_zod3.z.array(import_zod3.z.string()).optional().describe(
12983
+ 'Required if speaker_type is "role". Each value must be 35 characters or less.'
12984
+ ),
12985
+ speakers: import_zod3.z.array(
12986
+ import_zod3.z.object({
12987
+ role: import_zod3.z.string().optional().describe(
12988
+ 'The role of the speaker. Required when `speaker_type` is "role".'
12989
+ ),
12990
+ name: import_zod3.z.string().optional().describe(
12991
+ 'The name of the speaker. Required when `speaker_type` is "name".'
12992
+ ),
12993
+ description: import_zod3.z.string().optional().describe(
12994
+ "A description of the speaker to help the model identify them based on conversational context."
12995
+ )
12996
+ })
12997
+ ).optional().describe(
12998
+ "An array of speaker objects with metadata to improve identification accuracy. Each object should include a `role` or `name` (depending on `speaker_type`) and an optional `description` to help the model identify the speaker. You can also include any additional custom properties (e.g., `company`, `title`) to provide more context. Use this as an alternative to `known_values` when you want to provide additional context about each speaker."
12999
+ )
13000
+ })
13001
+ }).describe(
13002
+ "Request body for [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification)."
13003
+ )
13004
+ ).or(
13005
+ import_zod3.z.object({
13006
+ custom_formatting: import_zod3.z.object({
13007
+ date: import_zod3.z.string().optional().describe(
13008
+ 'Date format pattern (e.g., `"mm/dd/yyyy"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
13009
+ ),
13010
+ phone_number: import_zod3.z.string().optional().describe(
13011
+ 'Phone number format pattern (e.g., `"(xxx)xxx-xxxx"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
13012
+ ),
13013
+ email: import_zod3.z.string().optional().describe(
13014
+ 'Email format pattern (e.g., `"username@domain.com"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
13015
+ )
13016
+ })
13017
+ }).describe(
13018
+ "Request body for [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting)."
13019
+ )
13020
+ ).optional(),
13021
+ response: import_zod3.z.object({
13022
+ translation: import_zod3.z.object({
13023
+ status: import_zod3.z.string().optional()
13024
+ }).optional()
13025
+ }).or(
13026
+ import_zod3.z.object({
13027
+ speaker_identification: import_zod3.z.object({
13028
+ mapping: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional().describe(
13029
+ 'A mapping of the original generic speaker labels (e.g., "A", "B") to the identified speaker names or roles.'
13030
+ ),
13031
+ status: import_zod3.z.string().optional()
13032
+ }).optional()
13033
+ })
13034
+ ).or(
13035
+ import_zod3.z.object({
13036
+ custom_formatting: import_zod3.z.object({
13037
+ mapping: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional(),
13038
+ formatted_text: import_zod3.z.string().optional()
13039
+ }).optional()
13040
+ })
13041
+ ).optional()
13042
+ }).optional().describe(
13043
+ "Speech understanding tasks like [Translation](https://www.assemblyai.com/docs/speech-understanding/translation), [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification), and [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting). See the task-specific docs for available options and configuration.\n"
13044
+ ),
13045
+ status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).describe(
13046
+ "The status of your transcript. Possible values are queued, processing, completed, or error."
13047
+ ),
13048
+ summarization: import_zod3.z.boolean().describe(
13049
+ "Whether [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization) is enabled, either true or false. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details.\n\nNote: This parameter is only supported for the Universal-2 model.\n"
13050
+ ),
13051
+ summary: import_zod3.z.string().nullish().describe(
13052
+ "The generated summary of the media file, if [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization) is enabled. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details."
13053
+ ),
13054
+ summary_model: import_zod3.z.string().nullish().describe(
13055
+ "The Summarization model used to generate the summary,\nif [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization#summary-models) is enabled. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details.\n"
13056
+ ),
13057
+ summary_type: import_zod3.z.string().nullish().describe(
13058
+ "The type of summary generated, if [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization#summary-types) is enabled. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details."
13059
+ ),
13060
+ remove_audio_tags: import_zod3.z.enum(["all"]).describe(
13061
+ "Whether [audio event tags](https://www.assemblyai.com/docs/pre-recorded-audio/universal-3-pro#audio-event-tags) were removed from the transcript text.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n"
13062
+ ).or(import_zod3.z.null()).optional().describe(
13063
+ "Whether [audio event tags](https://www.assemblyai.com/docs/pre-recorded-audio/universal-3-pro#audio-event-tags) were removed from the transcript text.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n"
13064
+ ),
13065
+ temperature: import_zod3.z.number().nullish().describe(
13066
+ "The temperature that was used for the model's response. See the [Prompting Guide](https://www.assemblyai.com/docs/pre-recorded-audio/prompting) for more details.\n\nNote: This parameter can only be used with the Universal-3 Pro model.\n"
13067
+ ),
13068
+ text: import_zod3.z.string().nullish().describe("The textual transcript of your media file"),
13069
+ throttled: import_zod3.z.boolean().nullish().describe(
13070
+ "True while a request is throttled and false when a request is no longer throttled"
13071
+ ),
13072
+ utterances: import_zod3.z.array(
13073
+ import_zod3.z.object({
13074
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this utterance"),
13075
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, of the utterance in the audio file"),
13076
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, of the utterance in the audio file"),
13077
+ text: import_zod3.z.string().describe("The text for this utterance"),
13078
+ words: import_zod3.z.array(
13079
+ import_zod3.z.object({
13080
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
13081
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
13082
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
13083
+ text: import_zod3.z.string().describe("The text of the word"),
13084
+ channel: import_zod3.z.string().nullish().describe(
13085
+ "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13086
+ ),
13087
+ speaker: import_zod3.z.string().nullable().describe(
13088
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
13089
+ )
13090
+ })
13091
+ ).describe("The words in the utterance."),
13092
+ channel: import_zod3.z.string().nullish().describe(
13093
+ "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13094
+ ),
13095
+ speaker: import_zod3.z.string().describe(
13096
+ 'The speaker of this utterance, where each speaker is assigned a sequential capital letter - e.g. "A" for Speaker A, "B" for Speaker B, etc.'
13097
+ ),
13098
+ translated_texts: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional().describe(
13099
+ 'Translations keyed by language code (e.g., `{"es": "Texto traducido", "de": "\xDCbersetzter Text"}`). Only present when `match_original_utterance` is enabled with translation.'
13100
+ )
13101
+ })
13102
+ ).nullish().describe(
13103
+ "When multichannel or speaker_labels is enabled, a list of turn-by-turn utterance objects.\nSee [Speaker diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) and [Multichannel transcription](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) for more information.\n"
13104
+ ),
13105
+ webhook_auth: import_zod3.z.boolean().describe(
13106
+ "Whether [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) authentication details were provided"
13107
+ ),
13108
+ webhook_auth_header_name: import_zod3.z.string().nullish().describe(
13109
+ "The header name to be sent with the transcript completed or failed [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) requests"
13110
+ ),
13111
+ webhook_status_code: import_zod3.z.number().nullish().describe(
13112
+ "The status code we received from your server when delivering the transcript completed or failed [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) request, if a webhook URL was provided"
13113
+ ),
13114
+ webhook_url: import_zod3.z.string().nullish().describe(
13115
+ "The URL to which we send [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) requests.\n"
13116
+ ),
13117
+ words: import_zod3.z.array(
13118
+ import_zod3.z.object({
13119
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
13120
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
13121
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
13122
+ text: import_zod3.z.string().describe("The text of the word"),
13123
+ channel: import_zod3.z.string().nullish().describe(
13124
+ "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13125
+ ),
13126
+ speaker: import_zod3.z.string().nullable().describe(
13127
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
13128
+ )
13129
+ })
13130
+ ).nullish().describe(
13131
+ "An array of temporally-sequential word objects, one for each word in the transcript.\n"
13132
+ ),
13133
+ acoustic_model: import_zod3.z.string().describe("This parameter does not currently have any functionality attached to it."),
13134
+ custom_topics: import_zod3.z.boolean().nullish().describe("This parameter does not currently have any functionality attached to it."),
13135
+ language_model: import_zod3.z.string().describe("This parameter does not currently have any functionality attached to it."),
13136
+ speech_model: import_zod3.z.string().describe(
13137
+ "The speech model to use for the transcription. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models."
13138
+ ).or(import_zod3.z.null()).describe(
13139
+ "This parameter has been replaced with the `speech_models` parameter, learn more about the `speech_models` parameter [here](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model).\n"
13140
+ ),
13141
+ speed_boost: import_zod3.z.boolean().nullish().describe("This parameter does not currently have any functionality attached to it."),
13142
+ topics: import_zod3.z.array(import_zod3.z.string()).optional().describe("This parameter does not currently have any functionality attached to it."),
13143
+ translated_texts: import_zod3.z.object({
13144
+ language_code: import_zod3.z.string().optional().describe("Translated text for this language code")
13145
+ }).optional().describe(
13146
+ "Translated text keyed by language code. See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for more details."
13147
+ )
13148
+ }).describe("A transcript object");
13149
+ var listTranscriptsQueryLimitDefault = 10;
13150
+ var listTranscriptsQueryThrottledOnlyDefault = false;
13151
+ var listTranscriptsQueryParams = import_zod3.z.object({
13152
+ limit: import_zod3.z.number().default(listTranscriptsQueryLimitDefault).describe("Maximum amount of transcripts to retrieve"),
13153
+ status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).optional().describe("Filter by transcript status"),
13154
+ created_on: import_zod3.z.string().date().optional().describe("Only get transcripts created on this date"),
13155
+ before_id: import_zod3.z.string().uuid().optional().describe("Get transcripts that were created before this transcript ID"),
13156
+ after_id: import_zod3.z.string().uuid().optional().describe("Get transcripts that were created after this transcript ID"),
13157
+ throttled_only: import_zod3.z.boolean().optional().describe("Only get throttled transcripts, overrides the status filter")
13158
+ });
13159
+ var listTranscriptsResponse = import_zod3.z.object({
13160
+ page_details: import_zod3.z.object({
13161
+ limit: import_zod3.z.number().describe("The number of results this page is limited to"),
13162
+ result_count: import_zod3.z.number().describe("The actual number of results in the page"),
13163
+ current_url: import_zod3.z.string().describe("The URL used to retrieve the current page of transcripts"),
13164
+ prev_url: import_zod3.z.string().nullable().describe(
13165
+ "The URL to the next page of transcripts. The previous URL always points to a page with older transcripts."
13166
+ ),
13167
+ next_url: import_zod3.z.string().nullable().describe(
13168
+ "The URL to the next page of transcripts. The next URL always points to a page with newer transcripts."
13169
+ )
13170
+ }).describe(
13171
+ "Details of the transcript page. Transcripts are sorted from newest to oldest. The previous URL always points to a page with older transcripts."
13172
+ ),
13173
+ transcripts: import_zod3.z.array(
13174
+ import_zod3.z.object({
13175
+ id: import_zod3.z.string().uuid().describe("The unique identifier for the transcript"),
13176
+ resource_url: import_zod3.z.string().describe("The URL to retrieve the transcript"),
13177
+ status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).describe(
13178
+ "The status of your transcript. Possible values are queued, processing, completed, or error."
13179
+ ),
13180
+ created: import_zod3.z.string().datetime({}).describe("The date and time the transcript was created"),
13181
+ completed: import_zod3.z.string().datetime({}).optional().describe("The date and time the transcript was completed"),
13182
+ audio_url: import_zod3.z.string().describe("The URL to the audio file"),
13183
+ error: import_zod3.z.string().nullable().describe("Error message of why the transcript failed")
13184
+ })
13185
+ ).describe("An array of transcripts")
13186
+ }).describe(
13187
+ "A list of transcripts. Transcripts are sorted from newest to oldest. The previous URL always points to a page with older transcripts."
13188
+ );
13189
+ var getTranscriptParams = import_zod3.z.object({
13190
+ transcript_id: import_zod3.z.string().describe("ID of the transcript")
13191
+ });
13192
+ var getTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault = "auto";
13193
+ var getTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault = false;
13194
+ var getTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault = 0.3;
13195
+ var getTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault = false;
13196
+ var getTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault = true;
13197
+ var getTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault = false;
13198
+ var getTranscriptResponse = import_zod3.z.object({
13199
+ audio_channels: import_zod3.z.number().optional().describe(
13200
+ "The number of audio channels in the audio file. This is only present when [multichannel](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) is enabled."
13201
+ ),
13202
+ audio_duration: import_zod3.z.number().nullish().describe("The duration of this transcript object's media file, in seconds"),
13203
+ audio_end_at: import_zod3.z.number().nullish().describe(
13204
+ "The point in time, in milliseconds, in the file at which the transcription was terminated. See [Set the start and end of the transcript](https://www.assemblyai.com/docs/pre-recorded-audio/set-the-start-and-end-of-the-transcript) for more details."
13205
+ ),
13206
+ audio_start_from: import_zod3.z.number().nullish().describe(
13207
+ "The point in time, in milliseconds, in the file at which the transcription was started. See [Set the start and end of the transcript](https://www.assemblyai.com/docs/pre-recorded-audio/set-the-start-and-end-of-the-transcript) for more details."
13208
+ ),
13209
+ audio_url: import_zod3.z.string().describe("The URL of the media that was transcribed"),
13210
+ auto_chapters: import_zod3.z.boolean().nullish().describe(
13211
+ "Whether [Auto Chapters](https://www.assemblyai.com/docs/speech-understanding/auto-chapters) is enabled, can be true or false. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible chapter summaries. See the [updated Auto Chapters page](https://www.assemblyai.com/docs/speech-understanding/auto-chapters) for details.\n\nNote: This parameter is only supported for the Universal-2 model.\n"
13212
+ ),
13213
+ auto_highlights: import_zod3.z.boolean().describe(
13214
+ "Whether [Key Phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases) is enabled, either true or false"
13215
+ ),
13216
+ auto_highlights_result: import_zod3.z.object({
13217
+ status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
13218
+ results: import_zod3.z.array(
13219
+ import_zod3.z.object({
13220
+ count: import_zod3.z.number().describe("The total number of times the key phrase appears in the audio file"),
13221
+ rank: import_zod3.z.number().describe(
13222
+ "The total relevancy to the overall audio file of this key phrase - a greater number means more relevant"
13223
+ ),
13224
+ text: import_zod3.z.string().describe("The text itself of the key phrase"),
13225
+ timestamps: import_zod3.z.array(
13226
+ import_zod3.z.object({
13227
+ start: import_zod3.z.number().describe("The start time in milliseconds"),
13228
+ end: import_zod3.z.number().describe("The end time in milliseconds")
13229
+ }).describe("Timestamp containing a start and end property in milliseconds")
13230
+ ).describe("The timestamp of the of the key phrase")
13231
+ })
13232
+ ).describe("A temporally-sequential array of Key Phrases")
13233
+ }).describe(
13234
+ "An array of results for the Key Phrases model, if it is enabled.\nSee [Key phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases) for more information.\n"
13235
+ ).or(import_zod3.z.null()).optional().describe(
13236
+ "An array of results for the Key Phrases model, if it is enabled.\nSee [Key Phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases) for more information.\n"
13237
+ ),
13238
+ chapters: import_zod3.z.array(
13239
+ import_zod3.z.object({
13240
+ gist: import_zod3.z.string().describe(
13241
+ "An ultra-short summary (just a few words) of the content spoken in the chapter"
13242
+ ),
13243
+ headline: import_zod3.z.string().describe("A single sentence summary of the content spoken during the chapter"),
13244
+ summary: import_zod3.z.string().describe("A one paragraph summary of the content spoken during the chapter"),
13245
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter"),
13246
+ end: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter")
13247
+ }).describe("Chapter of the audio file")
13248
+ ).nullish().describe(
13249
+ "An array of temporally sequential chapters for the audio file. See [Auto Chapters](https://www.assemblyai.com/docs/speech-understanding/auto-chapters) for more information."
13250
+ ),
13251
+ confidence: import_zod3.z.number().nullish().describe(
13252
+ "The confidence score for the transcript, between 0.0 (low confidence) and 1.0 (high confidence)"
12453
13253
  ),
12454
13254
  content_safety: import_zod3.z.boolean().nullish().describe(
12455
- "Whether [Content Moderation](https://www.assemblyai.com/docs/models/content-moderation) is enabled, can be true or false"
13255
+ "Whether [Content Moderation](https://www.assemblyai.com/docs/content-moderation) is enabled, can be true or false"
12456
13256
  ),
12457
13257
  content_safety_labels: import_zod3.z.object({
12458
13258
  status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
@@ -12462,16 +13262,8 @@ var createTranscriptResponse = import_zod3.z.object({
12462
13262
  labels: import_zod3.z.array(
12463
13263
  import_zod3.z.object({
12464
13264
  label: import_zod3.z.string().describe("The label of the sensitive topic"),
12465
- confidence: import_zod3.z.number().min(
12466
- createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin
12467
- ).max(
12468
- createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax
12469
- ).describe("The confidence score for the topic being discussed, from 0 to 1"),
12470
- severity: import_zod3.z.number().min(
12471
- createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin
12472
- ).max(
12473
- createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax
12474
- ).describe("How severely the topic is discussed in the section, from 0 to 1")
13265
+ confidence: import_zod3.z.number().describe("The confidence score for the topic being discussed, from 0 to 1"),
13266
+ severity: import_zod3.z.number().describe("How severely the topic is discussed in the section, from 0 to 1")
12475
13267
  })
12476
13268
  ).describe(
12477
13269
  "An array of safety labels, one per sensitive topic that was detected in the section"
@@ -12484,128 +13276,39 @@ var createTranscriptResponse = import_zod3.z.object({
12484
13276
  }).describe("Timestamp containing a start and end property in milliseconds")
12485
13277
  })
12486
13278
  ).describe("An array of results for the Content Moderation model"),
12487
- summary: import_zod3.z.record(
12488
- import_zod3.z.string(),
12489
- import_zod3.z.number().min(createTranscriptResponseContentSafetyLabelsSummaryMinOne).max(createTranscriptResponseContentSafetyLabelsSummaryMaxOne).describe(
12490
- 'A confidence score for the presence of the sensitive topic "topic" across the entire audio file'
12491
- )
12492
- ).describe(
13279
+ summary: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.number()).describe(
12493
13280
  "A summary of the Content Moderation confidence results for the entire audio file"
12494
13281
  ),
12495
13282
  severity_score_summary: import_zod3.z.record(
12496
13283
  import_zod3.z.string(),
12497
13284
  import_zod3.z.object({
12498
- low: import_zod3.z.number().min(createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin).max(createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax),
12499
- medium: import_zod3.z.number().min(createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin).max(createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax),
12500
- high: import_zod3.z.number().min(createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin).max(createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax)
13285
+ low: import_zod3.z.number(),
13286
+ medium: import_zod3.z.number(),
13287
+ high: import_zod3.z.number()
12501
13288
  })
12502
13289
  ).describe(
12503
13290
  "A summary of the Content Moderation severity results for the entire audio file"
12504
13291
  )
12505
13292
  }).describe(
12506
- "An array of results for the Content Moderation model, if it is enabled.\nSee [Content moderation](https://www.assemblyai.com/docs/models/content-moderation) for more information.\n"
13293
+ "An array of results for the Content Moderation model, if it is enabled.\nSee [Content moderation](https://www.assemblyai.com/docs/content-moderation) for more information.\n"
12507
13294
  ).or(import_zod3.z.null()).optional().describe(
12508
- "An array of results for the Content Moderation model, if it is enabled.\nSee [Content moderation](https://www.assemblyai.com/docs/models/content-moderation) for more information.\n"
12509
- ),
12510
- iab_categories: import_zod3.z.boolean().nullish().describe(
12511
- "Whether [Topic Detection](https://www.assemblyai.com/docs/models/topic-detection) is enabled, can be true or false"
12512
- ),
12513
- iab_categories_result: import_zod3.z.object({
12514
- status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
12515
- results: import_zod3.z.array(
12516
- import_zod3.z.object({
12517
- text: import_zod3.z.string().describe("The text in the transcript in which a detected topic occurs"),
12518
- labels: import_zod3.z.array(
12519
- import_zod3.z.object({
12520
- relevance: import_zod3.z.number().min(
12521
- createTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin
12522
- ).max(
12523
- createTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax
12524
- ).describe("How relevant the detected topic is of a detected topic"),
12525
- label: import_zod3.z.string().describe(
12526
- "The IAB taxonomical label for the label of the detected topic, where > denotes supertopic/subtopic relationship"
12527
- )
12528
- })
12529
- ).optional().describe("An array of detected topics in the text"),
12530
- timestamp: import_zod3.z.object({
12531
- start: import_zod3.z.number().describe("The start time in milliseconds"),
12532
- end: import_zod3.z.number().describe("The end time in milliseconds")
12533
- }).optional().describe("Timestamp containing a start and end property in milliseconds")
12534
- }).describe("The result of the topic detection model")
12535
- ).describe("An array of results for the Topic Detection model"),
12536
- summary: import_zod3.z.record(
12537
- import_zod3.z.string(),
12538
- import_zod3.z.number().min(createTranscriptResponseIabCategoriesResultSummaryMinOne).max(createTranscriptResponseIabCategoriesResultSummaryMaxOne)
12539
- ).describe("The overall relevance of topic to the entire audio file")
12540
- }).describe(
12541
- "The result of the Topic Detection model, if it is enabled.\nSee [Topic Detection](https://www.assemblyai.com/docs/models/topic-detection) for more information.\n"
12542
- ).or(import_zod3.z.null()).optional().describe(
12543
- "The result of the Topic Detection model, if it is enabled.\nSee [Topic Detection](https://www.assemblyai.com/docs/models/topic-detection) for more information.\n"
13295
+ "An array of results for the Content Moderation model, if it is enabled.\nSee [Content moderation](https://www.assemblyai.com/docs/content-moderation) for more information.\n"
12544
13296
  ),
12545
13297
  custom_spelling: import_zod3.z.array(
12546
13298
  import_zod3.z.object({
12547
- from: import_zod3.z.array(import_zod3.z.string().describe("Word or phrase to replace")).describe("Words or phrases to replace"),
13299
+ from: import_zod3.z.array(import_zod3.z.string()).describe("Words or phrases to replace"),
12548
13300
  to: import_zod3.z.string().describe("Word to replace with")
12549
13301
  }).describe(
12550
13302
  "Object containing words or phrases to replace, and the word or phrase to replace with"
12551
13303
  )
12552
- ).nullish().describe("Customize how words are spelled and formatted using to and from values"),
12553
- keyterms_prompt: import_zod3.z.array(import_zod3.z.string()).optional().describe(
12554
- "Improve accuracy with up to 1000 domain-specific words or phrases (maximum 6 words per phrase).\n"
12555
- ),
12556
- prompt: import_zod3.z.string().optional().describe("This parameter does not currently have any functionality attached to it."),
12557
- auto_chapters: import_zod3.z.boolean().nullish().describe(
12558
- "Whether [Auto Chapters](https://www.assemblyai.com/docs/models/auto-chapters) is enabled, can be true or false"
12559
- ),
12560
- chapters: import_zod3.z.array(
12561
- import_zod3.z.object({
12562
- gist: import_zod3.z.string().describe(
12563
- "An ultra-short summary (just a few words) of the content spoken in the chapter"
12564
- ),
12565
- headline: import_zod3.z.string().describe("A single sentence summary of the content spoken during the chapter"),
12566
- summary: import_zod3.z.string().describe("A one paragraph summary of the content spoken during the chapter"),
12567
- start: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter"),
12568
- end: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter")
12569
- }).describe("Chapter of the audio file")
12570
- ).nullish().describe("An array of temporally sequential chapters for the audio file"),
12571
- summarization: import_zod3.z.boolean().describe(
12572
- "Whether [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled, either true or false"
12573
- ),
12574
- summary_type: import_zod3.z.string().nullish().describe(
12575
- "The type of summary generated, if [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled"
12576
- ),
12577
- summary_model: import_zod3.z.string().nullish().describe(
12578
- "The Summarization model used to generate the summary,\nif [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled\n"
12579
- ),
12580
- summary: import_zod3.z.string().nullish().describe(
12581
- "The generated summary of the media file, if [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled"
12582
- ),
12583
- custom_topics: import_zod3.z.boolean().nullish().describe("Whether custom topics is enabled, either true or false"),
12584
- topics: import_zod3.z.array(import_zod3.z.string()).optional().describe("The list of custom topics provided if custom topics is enabled"),
12585
- sentiment_analysis: import_zod3.z.boolean().nullish().describe(
12586
- "Whether [Sentiment Analysis](https://www.assemblyai.com/docs/models/sentiment-analysis) is enabled, can be true or false"
12587
- ),
12588
- sentiment_analysis_results: import_zod3.z.array(
12589
- import_zod3.z.object({
12590
- text: import_zod3.z.string().describe("The transcript of the sentence"),
12591
- start: import_zod3.z.number().describe("The starting time, in milliseconds, of the sentence"),
12592
- end: import_zod3.z.number().describe("The ending time, in milliseconds, of the sentence"),
12593
- sentiment: import_zod3.z.enum(["POSITIVE", "NEUTRAL", "NEGATIVE"]),
12594
- confidence: import_zod3.z.number().min(createTranscriptResponseSentimentAnalysisResultsItemConfidenceMin).max(createTranscriptResponseSentimentAnalysisResultsItemConfidenceMax).describe(
12595
- "The confidence score for the detected sentiment of the sentence, from 0 to 1"
12596
- ),
12597
- channel: import_zod3.z.string().nullish().describe(
12598
- "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
12599
- ),
12600
- speaker: import_zod3.z.string().nullable().describe(
12601
- "The speaker of the sentence if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
12602
- )
12603
- }).describe("The result of the Sentiment Analysis model")
12604
13304
  ).nullish().describe(
12605
- "An array of results for the Sentiment Analysis model, if it is enabled.\nSee [Sentiment Analysis](https://www.assemblyai.com/docs/models/sentiment-analysis) for more information.\n"
13305
+ "Customize how words are spelled and formatted using to and from values. See [Custom Spelling](https://www.assemblyai.com/docs/pre-recorded-audio/custom-spelling) for more details."
12606
13306
  ),
12607
- entity_detection: import_zod3.z.boolean().nullish().describe(
12608
- "Whether [Entity Detection](https://www.assemblyai.com/docs/models/entity-detection) is enabled, can be true or false"
13307
+ disfluencies: import_zod3.z.boolean().nullish().describe(
13308
+ 'Transcribe [Filler Words](https://www.assemblyai.com/docs/pre-recorded-audio/filler-words), like "umm", in your media file; can be true or false'
13309
+ ),
13310
+ domain: import_zod3.z.string().nullish().describe(
13311
+ 'The domain-specific model applied to the transcript. When set to `"medical-v1"`, [Medical Mode](https://www.assemblyai.com/docs/pre-recorded-audio/medical-mode) was used to improve accuracy for medical terminology.\n'
12609
13312
  ),
12610
13313
  entities: import_zod3.z.array(
12611
13314
  import_zod3.z.object({
@@ -12664,108 +13367,49 @@ var createTranscriptResponse = import_zod3.z.object({
12664
13367
  )
12665
13368
  }).describe("A detected entity")
12666
13369
  ).nullish().describe(
12667
- "An array of results for the Entity Detection model, if it is enabled.\nSee [Entity detection](https://www.assemblyai.com/docs/models/entity-detection) for more information.\n"
12668
- ),
12669
- speech_threshold: import_zod3.z.number().min(createTranscriptResponseSpeechThresholdMin).max(createTranscriptResponseSpeechThresholdMax).nullish().describe(
12670
- "Defaults to null. Reject audio files that contain less than this fraction of speech.\nValid values are in the range [0, 1] inclusive.\n"
13370
+ "An array of results for the Entity Detection model, if it is enabled.\nSee [Entity detection](https://www.assemblyai.com/docs/speech-understanding/entity-detection) for more information.\n"
12671
13371
  ),
12672
- throttled: import_zod3.z.boolean().nullish().describe(
12673
- "True while a request is throttled and false when a request is no longer throttled"
13372
+ entity_detection: import_zod3.z.boolean().nullish().describe(
13373
+ "Whether [Entity Detection](https://www.assemblyai.com/docs/speech-understanding/entity-detection) is enabled, can be true or false"
12674
13374
  ),
12675
13375
  error: import_zod3.z.string().optional().describe("Error message of why the transcript failed"),
12676
- language_model: import_zod3.z.string().describe("The language model that was used for the transcript"),
12677
- acoustic_model: import_zod3.z.string().describe("The acoustic model that was used for the transcript")
12678
- }).describe("A transcript object");
12679
- var listTranscriptsQueryLimitDefault = 10;
12680
- var listTranscriptsQueryLimitMax = 200;
12681
- var listTranscriptsQueryThrottledOnlyDefault = false;
12682
- var listTranscriptsQueryParams = import_zod3.z.object({
12683
- limit: import_zod3.z.number().min(1).max(listTranscriptsQueryLimitMax).default(listTranscriptsQueryLimitDefault).describe("Maximum amount of transcripts to retrieve"),
12684
- status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).optional().describe("Filter by transcript status"),
12685
- created_on: import_zod3.z.string().date().optional().describe("Only get transcripts created on this date"),
12686
- before_id: import_zod3.z.string().uuid().optional().describe("Get transcripts that were created before this transcript ID"),
12687
- after_id: import_zod3.z.string().uuid().optional().describe("Get transcripts that were created after this transcript ID"),
12688
- throttled_only: import_zod3.z.boolean().optional().describe("Only get throttled transcripts, overrides the status filter")
12689
- });
12690
- var listTranscriptsResponseTranscriptsItemCreatedRegExp = new RegExp(
12691
- "^(?:(\\d{4}-\\d{2}-\\d{2})T(\\d{2}:\\d{2}:\\d{2}(?:\\.\\d+)?))$"
12692
- );
12693
- var listTranscriptsResponseTranscriptsItemCompletedRegExp = new RegExp(
12694
- "^(?:(\\d{4}-\\d{2}-\\d{2})T(\\d{2}:\\d{2}:\\d{2}(?:\\.\\d+)?))$"
12695
- );
12696
- var listTranscriptsResponse = import_zod3.z.object({
12697
- page_details: import_zod3.z.object({
12698
- limit: import_zod3.z.number().describe("The number of results this page is limited to"),
12699
- result_count: import_zod3.z.number().describe("The actual number of results in the page"),
12700
- current_url: import_zod3.z.string().describe("The URL used to retrieve the current page of transcripts"),
12701
- prev_url: import_zod3.z.string().nullable().describe(
12702
- "The URL to the next page of transcripts. The previous URL always points to a page with older transcripts."
12703
- ),
12704
- next_url: import_zod3.z.string().nullable().describe(
12705
- "The URL to the next page of transcripts. The next URL always points to a page with newer transcripts."
12706
- )
13376
+ filter_profanity: import_zod3.z.boolean().nullish().describe(
13377
+ "Whether [Profanity Filtering](https://www.assemblyai.com/docs/profanity-filtering) is enabled, either true or false"
13378
+ ),
13379
+ format_text: import_zod3.z.boolean().nullish().describe(
13380
+ "Whether [Text Formatting](https://www.assemblyai.com/docs/pre-recorded-audio) is enabled, either true or false"
13381
+ ),
13382
+ iab_categories: import_zod3.z.boolean().nullish().describe(
13383
+ "Whether [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection) is enabled, can be true or false"
13384
+ ),
13385
+ iab_categories_result: import_zod3.z.object({
13386
+ status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
13387
+ results: import_zod3.z.array(
13388
+ import_zod3.z.object({
13389
+ text: import_zod3.z.string().describe("The text in the transcript in which a detected topic occurs"),
13390
+ labels: import_zod3.z.array(
13391
+ import_zod3.z.object({
13392
+ relevance: import_zod3.z.number().describe("How relevant the detected topic is of a detected topic"),
13393
+ label: import_zod3.z.string().describe(
13394
+ "The IAB taxonomical label for the label of the detected topic, where > denotes supertopic/subtopic relationship"
13395
+ )
13396
+ })
13397
+ ).optional().describe("An array of detected topics in the text"),
13398
+ timestamp: import_zod3.z.object({
13399
+ start: import_zod3.z.number().describe("The start time in milliseconds"),
13400
+ end: import_zod3.z.number().describe("The end time in milliseconds")
13401
+ }).optional().describe("Timestamp containing a start and end property in milliseconds")
13402
+ }).describe("The result of the topic detection model")
13403
+ ).describe("An array of results for the Topic Detection model"),
13404
+ summary: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.number()).describe("The overall relevance of topic to the entire audio file")
12707
13405
  }).describe(
12708
- "Details of the transcript page. Transcripts are sorted from newest to oldest. The previous URL always points to a page with older transcripts."
13406
+ "The result of the Topic Detection model, if it is enabled.\nSee [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection) for more information.\n"
13407
+ ).or(import_zod3.z.null()).optional().describe(
13408
+ "The result of the Topic Detection model, if it is enabled.\nSee [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection) for more information.\n"
12709
13409
  ),
12710
- transcripts: import_zod3.z.array(
12711
- import_zod3.z.object({
12712
- id: import_zod3.z.string().uuid().describe("The unique identifier for the transcript"),
12713
- resource_url: import_zod3.z.string().describe("The URL to retrieve the transcript"),
12714
- status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).describe(
12715
- "The status of your transcript. Possible values are queued, processing, completed, or error."
12716
- ),
12717
- created: import_zod3.z.string().regex(listTranscriptsResponseTranscriptsItemCreatedRegExp).describe("The date and time the transcript was created"),
12718
- completed: import_zod3.z.string().regex(listTranscriptsResponseTranscriptsItemCompletedRegExp).nullable().describe("The date and time the transcript was completed"),
12719
- audio_url: import_zod3.z.string().describe("The URL to the audio file"),
12720
- error: import_zod3.z.string().nullable().describe("Error message of why the transcript failed")
12721
- })
12722
- ).describe("An array of transcripts")
12723
- }).describe(
12724
- "A list of transcripts. Transcripts are sorted from newest to oldest. The previous URL always points to a page with older transcripts."
12725
- );
12726
- var getTranscriptParams = import_zod3.z.object({
12727
- transcript_id: import_zod3.z.string().describe("ID of the transcript")
12728
- });
12729
- var getTranscriptResponseLanguageConfidenceThresholdMin = 0;
12730
- var getTranscriptResponseLanguageConfidenceThresholdMax = 1;
12731
- var getTranscriptResponseLanguageConfidenceMin = 0;
12732
- var getTranscriptResponseLanguageConfidenceMax = 1;
12733
- var getTranscriptResponseSpeechModelDefault = null;
12734
- var getTranscriptResponseWordsItemConfidenceMin = 0;
12735
- var getTranscriptResponseWordsItemConfidenceMax = 1;
12736
- var getTranscriptResponseUtterancesItemConfidenceMin = 0;
12737
- var getTranscriptResponseUtterancesItemConfidenceMax = 1;
12738
- var getTranscriptResponseUtterancesItemWordsItemConfidenceMin = 0;
12739
- var getTranscriptResponseUtterancesItemWordsItemConfidenceMax = 1;
12740
- var getTranscriptResponseConfidenceMin = 0;
12741
- var getTranscriptResponseConfidenceMax = 1;
12742
- var getTranscriptResponseAutoHighlightsResultResultsItemRankMin = 0;
12743
- var getTranscriptResponseAutoHighlightsResultResultsItemRankMax = 1;
12744
- var getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin = 0;
12745
- var getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax = 1;
12746
- var getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin = 0;
12747
- var getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax = 1;
12748
- var getTranscriptResponseContentSafetyLabelsSummaryMinOne = 0;
12749
- var getTranscriptResponseContentSafetyLabelsSummaryMaxOne = 1;
12750
- var getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin = 0;
12751
- var getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax = 1;
12752
- var getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin = 0;
12753
- var getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax = 1;
12754
- var getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin = 0;
12755
- var getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax = 1;
12756
- var getTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin = 0;
12757
- var getTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax = 1;
12758
- var getTranscriptResponseIabCategoriesResultSummaryMinOne = 0;
12759
- var getTranscriptResponseIabCategoriesResultSummaryMaxOne = 1;
12760
- var getTranscriptResponseSentimentAnalysisResultsItemConfidenceMin = 0;
12761
- var getTranscriptResponseSentimentAnalysisResultsItemConfidenceMax = 1;
12762
- var getTranscriptResponseSpeechThresholdMin = 0;
12763
- var getTranscriptResponseSpeechThresholdMax = 1;
12764
- var getTranscriptResponse = import_zod3.z.object({
12765
13410
  id: import_zod3.z.string().uuid().describe("The unique identifier of your transcript"),
12766
- audio_url: import_zod3.z.string().describe("The URL of the media that was transcribed"),
12767
- status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).describe(
12768
- "The status of your transcript. Possible values are queued, processing, completed, or error."
13411
+ keyterms_prompt: import_zod3.z.array(import_zod3.z.string()).optional().describe(
13412
+ "Improve accuracy with up to 200 (for Universal-2) or 1000 (for Universal-3 Pro) domain-specific words or phrases (maximum 6 words per phrase). See [Keyterms Prompting](https://www.assemblyai.com/docs/pre-recorded-audio/keyterms-prompting) for more details.\n"
12769
13413
  ),
12770
13414
  language_code: import_zod3.z.enum([
12771
13415
  "en",
@@ -12870,138 +13514,175 @@ var getTranscriptResponse = import_zod3.z.object({
12870
13514
  "cy",
12871
13515
  "yi",
12872
13516
  "yo"
12873
- ]).describe(
12874
- "The language of your audio file. Possible values are found in [Supported Languages](https://www.assemblyai.com/docs/concepts/supported-languages).\nThe default value is 'en_us'.\n"
12875
- ).or(import_zod3.z.string()).optional().describe(
12876
- "The language of your audio file.\nPossible values are found in [Supported Languages](https://www.assemblyai.com/docs/concepts/supported-languages).\nThe default value is 'en_us'.\n"
12877
- ),
12878
- language_detection: import_zod3.z.boolean().nullish().describe(
12879
- "Whether [Automatic language detection](https://www.assemblyai.com/docs/models/speech-recognition#automatic-language-detection) is enabled, either true or false"
12880
- ),
12881
- language_confidence_threshold: import_zod3.z.number().min(getTranscriptResponseLanguageConfidenceThresholdMin).max(getTranscriptResponseLanguageConfidenceThresholdMax).nullable().describe(
12882
- "The confidence threshold for the automatically detected language.\nAn error will be returned if the language confidence is below this threshold.\n"
12883
- ),
12884
- language_confidence: import_zod3.z.number().min(getTranscriptResponseLanguageConfidenceMin).max(getTranscriptResponseLanguageConfidenceMax).nullable().describe(
12885
- "The confidence score for the detected language, between 0.0 (low confidence) and 1.0 (high confidence)"
12886
- ),
12887
- speech_model: import_zod3.z.enum(["best", "slam-1", "universal"]).describe("The speech model to use for the transcription.").or(import_zod3.z.null()).describe(
12888
- "The speech model used for the transcription. When `null`, the default model is used."
12889
- ),
12890
- text: import_zod3.z.string().nullish().describe("The textual transcript of your media file"),
12891
- words: import_zod3.z.array(
12892
- import_zod3.z.object({
12893
- confidence: import_zod3.z.number().min(getTranscriptResponseWordsItemConfidenceMin).max(getTranscriptResponseWordsItemConfidenceMax).describe("The confidence score for the transcript of this word"),
12894
- start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
12895
- end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
12896
- text: import_zod3.z.string().describe("The text of the word"),
12897
- channel: import_zod3.z.string().nullish().describe(
12898
- "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
12899
- ),
12900
- speaker: import_zod3.z.string().nullable().describe(
12901
- "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
12902
- )
12903
- })
12904
- ).nullish().describe(
12905
- "An array of temporally-sequential word objects, one for each word in the transcript.\nSee [Speech recognition](https://www.assemblyai.com/docs/models/speech-recognition) for more information.\n"
13517
+ ]).optional().describe(
13518
+ "The language of your audio file. Possible values are found in [Supported Languages](https://www.assemblyai.com/docs/pre-recorded-audio/supported-languages).\nThe default value is 'en_us'.\n"
12906
13519
  ),
12907
- utterances: import_zod3.z.array(
12908
- import_zod3.z.object({
12909
- confidence: import_zod3.z.number().min(getTranscriptResponseUtterancesItemConfidenceMin).max(getTranscriptResponseUtterancesItemConfidenceMax).describe("The confidence score for the transcript of this utterance"),
12910
- start: import_zod3.z.number().describe("The starting time, in milliseconds, of the utterance in the audio file"),
12911
- end: import_zod3.z.number().describe("The ending time, in milliseconds, of the utterance in the audio file"),
12912
- text: import_zod3.z.string().describe("The text for this utterance"),
12913
- words: import_zod3.z.array(
12914
- import_zod3.z.object({
12915
- confidence: import_zod3.z.number().min(getTranscriptResponseUtterancesItemWordsItemConfidenceMin).max(getTranscriptResponseUtterancesItemWordsItemConfidenceMax).describe("The confidence score for the transcript of this word"),
12916
- start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
12917
- end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
12918
- text: import_zod3.z.string().describe("The text of the word"),
12919
- channel: import_zod3.z.string().nullish().describe(
12920
- "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
12921
- ),
12922
- speaker: import_zod3.z.string().nullable().describe(
12923
- "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
12924
- )
12925
- })
12926
- ).describe("The words in the utterance."),
12927
- channel: import_zod3.z.string().nullish().describe(
12928
- "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
12929
- ),
12930
- speaker: import_zod3.z.string().describe(
12931
- 'The speaker of this utterance, where each speaker is assigned a sequential capital letter - e.g. "A" for Speaker A, "B" for Speaker B, etc.'
12932
- )
12933
- })
13520
+ language_codes: import_zod3.z.array(
13521
+ import_zod3.z.enum([
13522
+ "en",
13523
+ "en_au",
13524
+ "en_uk",
13525
+ "en_us",
13526
+ "es",
13527
+ "fr",
13528
+ "de",
13529
+ "it",
13530
+ "pt",
13531
+ "nl",
13532
+ "af",
13533
+ "sq",
13534
+ "am",
13535
+ "ar",
13536
+ "hy",
13537
+ "as",
13538
+ "az",
13539
+ "ba",
13540
+ "eu",
13541
+ "be",
13542
+ "bn",
13543
+ "bs",
13544
+ "br",
13545
+ "bg",
13546
+ "my",
13547
+ "ca",
13548
+ "zh",
13549
+ "hr",
13550
+ "cs",
13551
+ "da",
13552
+ "et",
13553
+ "fo",
13554
+ "fi",
13555
+ "gl",
13556
+ "ka",
13557
+ "el",
13558
+ "gu",
13559
+ "ht",
13560
+ "ha",
13561
+ "haw",
13562
+ "he",
13563
+ "hi",
13564
+ "hu",
13565
+ "is",
13566
+ "id",
13567
+ "ja",
13568
+ "jw",
13569
+ "kn",
13570
+ "kk",
13571
+ "km",
13572
+ "ko",
13573
+ "lo",
13574
+ "la",
13575
+ "lv",
13576
+ "ln",
13577
+ "lt",
13578
+ "lb",
13579
+ "mk",
13580
+ "mg",
13581
+ "ms",
13582
+ "ml",
13583
+ "mt",
13584
+ "mi",
13585
+ "mr",
13586
+ "mn",
13587
+ "ne",
13588
+ "no",
13589
+ "nn",
13590
+ "oc",
13591
+ "pa",
13592
+ "ps",
13593
+ "fa",
13594
+ "pl",
13595
+ "ro",
13596
+ "ru",
13597
+ "sa",
13598
+ "sr",
13599
+ "sn",
13600
+ "sd",
13601
+ "si",
13602
+ "sk",
13603
+ "sl",
13604
+ "so",
13605
+ "su",
13606
+ "sw",
13607
+ "sv",
13608
+ "tl",
13609
+ "tg",
13610
+ "ta",
13611
+ "tt",
13612
+ "te",
13613
+ "th",
13614
+ "bo",
13615
+ "tr",
13616
+ "tk",
13617
+ "uk",
13618
+ "ur",
13619
+ "uz",
13620
+ "vi",
13621
+ "cy",
13622
+ "yi",
13623
+ "yo"
13624
+ ]).describe(
13625
+ "The language of your audio file. Possible values are found in [Supported Languages](https://www.assemblyai.com/docs/pre-recorded-audio/supported-languages).\nThe default value is 'en_us'.\n"
13626
+ )
12934
13627
  ).nullish().describe(
12935
- "When multichannel or speaker_labels is enabled, a list of turn-by-turn utterance objects.\nSee [Speaker diarization](https://www.assemblyai.com/docs/speech-to-text/speaker-diarization) and [Multichannel transcription](https://www.assemblyai.com/docs/speech-to-text/speech-recognition#multichannel-transcription) for more information.\n"
13628
+ "The language codes of your audio file. Used for [Code switching](/docs/speech-to-text/pre-recorded-audio/code-switching)\nOne of the values specified must be `en`.\n"
12936
13629
  ),
12937
- confidence: import_zod3.z.number().min(getTranscriptResponseConfidenceMin).max(getTranscriptResponseConfidenceMax).nullish().describe(
12938
- "The confidence score for the transcript, between 0.0 (low confidence) and 1.0 (high confidence)"
13630
+ language_confidence: import_zod3.z.number().nullable().describe(
13631
+ "The confidence score for the detected language, between 0.0 (low confidence) and 1.0 (high confidence). See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details."
12939
13632
  ),
12940
- audio_duration: import_zod3.z.number().nullish().describe("The duration of this transcript object's media file, in seconds"),
12941
- punctuate: import_zod3.z.boolean().nullish().describe("Whether Automatic Punctuation is enabled, either true or false"),
12942
- format_text: import_zod3.z.boolean().nullish().describe("Whether Text Formatting is enabled, either true or false"),
12943
- disfluencies: import_zod3.z.boolean().nullish().describe('Transcribe Filler Words, like "umm", in your media file; can be true or false'),
12944
- multichannel: import_zod3.z.boolean().nullish().describe(
12945
- "Whether [Multichannel transcription](https://www.assemblyai.com/docs/models/speech-recognition#multichannel-transcription) was enabled in the transcription request, either true or false"
12946
- ),
12947
- audio_channels: import_zod3.z.number().optional().describe(
12948
- "The number of audio channels in the audio file. This is only present when multichannel is enabled."
12949
- ),
12950
- webhook_url: import_zod3.z.string().nullish().describe(
12951
- "The URL to which we send webhook requests.\nWe sends two different types of webhook requests.\nOne request when a transcript is completed or failed, and one request when the redacted audio is ready if redact_pii_audio is enabled.\n"
12952
- ),
12953
- webhook_status_code: import_zod3.z.number().nullish().describe(
12954
- "The status code we received from your server when delivering the transcript completed or failed webhook request, if a webhook URL was provided"
13633
+ language_confidence_threshold: import_zod3.z.number().nullable().describe(
13634
+ "The confidence threshold for the automatically detected language.\nAn error will be returned if the language confidence is below this threshold.\nSee [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.\n"
12955
13635
  ),
12956
- webhook_auth: import_zod3.z.boolean().describe("Whether webhook authentication details were provided"),
12957
- webhook_auth_header_name: import_zod3.z.string().nullish().describe(
12958
- "The header name to be sent with the transcript completed or failed webhook requests"
13636
+ language_detection: import_zod3.z.boolean().nullish().describe(
13637
+ "Whether [Automatic language detection](/docs/pre-recorded-audio/automatic-language-detection) is enabled, either true or false"
12959
13638
  ),
12960
- speed_boost: import_zod3.z.boolean().nullish().describe("Whether speed boost is enabled"),
12961
- auto_highlights: import_zod3.z.boolean().describe("Whether Key Phrases is enabled, either true or false"),
12962
- auto_highlights_result: import_zod3.z.object({
12963
- status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
12964
- results: import_zod3.z.array(
12965
- import_zod3.z.object({
12966
- count: import_zod3.z.number().describe("The total number of times the key phrase appears in the audio file"),
12967
- rank: import_zod3.z.number().min(getTranscriptResponseAutoHighlightsResultResultsItemRankMin).max(getTranscriptResponseAutoHighlightsResultResultsItemRankMax).describe(
12968
- "The total relevancy to the overall audio file of this key phrase - a greater number means more relevant"
12969
- ),
12970
- text: import_zod3.z.string().describe("The text itself of the key phrase"),
12971
- timestamps: import_zod3.z.array(
12972
- import_zod3.z.object({
12973
- start: import_zod3.z.number().describe("The start time in milliseconds"),
12974
- end: import_zod3.z.number().describe("The end time in milliseconds")
12975
- }).describe("Timestamp containing a start and end property in milliseconds")
12976
- ).describe("The timestamp of the of the key phrase")
12977
- })
12978
- ).describe("A temporally-sequential array of Key Phrases")
12979
- }).describe(
12980
- "An array of results for the Key Phrases model, if it is enabled.\nSee [Key phrases](https://www.assemblyai.com/docs/models/key-phrases) for more information.\n"
12981
- ).or(import_zod3.z.null()).optional().describe(
12982
- "An array of results for the Key Phrases model, if it is enabled.\nSee [Key Phrases](https://www.assemblyai.com/docs/models/key-phrases) for more information.\n"
13639
+ language_detection_options: import_zod3.z.object({
13640
+ expected_languages: import_zod3.z.array(import_zod3.z.string()).optional().describe(
13641
+ 'List of languages expected in the audio file. Defaults to `["all"]` when unspecified. See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.'
13642
+ ),
13643
+ fallback_language: import_zod3.z.string().default(getTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault).describe(
13644
+ 'If the detected language of the audio file is not in the list of expected languages, the `fallback_language` is used. Specify `["auto"]` to let our model choose the fallback language from `expected_languages` with the highest confidence score. See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.\n'
13645
+ ),
13646
+ code_switching: import_zod3.z.boolean().optional().describe(
13647
+ "Whether [code switching](/docs/speech-to-text/pre-recorded-audio/code-switching) should be detected.\n"
13648
+ ),
13649
+ code_switching_confidence_threshold: import_zod3.z.number().default(
13650
+ getTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault
13651
+ ).describe(
13652
+ "The confidence threshold for [code switching](/docs/speech-to-text/pre-recorded-audio/code-switching) detection. If the code switching confidence is below this threshold, the transcript will be processed in the language with the highest `language_detection_confidence` score.\n"
13653
+ )
13654
+ }).optional().describe(
13655
+ "Specify options for [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection)."
12983
13656
  ),
12984
- audio_start_from: import_zod3.z.number().nullish().describe(
12985
- "The point in time, in milliseconds, in the file at which the transcription was started"
13657
+ multichannel: import_zod3.z.boolean().nullish().describe(
13658
+ "Whether [Multichannel transcription](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) was enabled in the transcription request, either true or false"
12986
13659
  ),
12987
- audio_end_at: import_zod3.z.number().nullish().describe(
12988
- "The point in time, in milliseconds, in the file at which the transcription was terminated"
13660
+ prompt: import_zod3.z.string().optional().describe(
13661
+ "Provide natural language prompting of up to 1,500 words of contextual information to the model. See the [Prompting Guide](https://www.assemblyai.com/docs/pre-recorded-audio/prompting) for best practices.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n"
12989
13662
  ),
12990
- word_boost: import_zod3.z.array(import_zod3.z.string()).optional().describe("The list of custom vocabulary to boost transcription probability for"),
12991
- boost_param: import_zod3.z.string().nullish().describe("The word boost parameter value"),
12992
- filter_profanity: import_zod3.z.boolean().nullish().describe(
12993
- "Whether [Profanity Filtering](https://www.assemblyai.com/docs/models/speech-recognition#profanity-filtering) is enabled, either true or false"
13663
+ punctuate: import_zod3.z.boolean().nullish().describe(
13664
+ "Whether [Automatic Punctuation](https://www.assemblyai.com/docs/pre-recorded-audio) is enabled, either true or false"
12994
13665
  ),
12995
13666
  redact_pii: import_zod3.z.boolean().describe(
12996
- "Whether [PII Redaction](https://www.assemblyai.com/docs/models/pii-redaction) is enabled, either true or false"
13667
+ "Whether [PII Redaction](https://www.assemblyai.com/docs/pii-redaction) is enabled, either true or false"
12997
13668
  ),
12998
13669
  redact_pii_audio: import_zod3.z.boolean().nullish().describe(
12999
- "Whether a redacted version of the audio file was generated,\neither true or false. See [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more information.\n"
13670
+ "Whether a redacted version of the audio file was generated,\neither true or false. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more information.\n"
13671
+ ),
13672
+ redact_pii_audio_options: import_zod3.z.object({
13673
+ return_redacted_no_speech_audio: import_zod3.z.boolean().optional().describe(
13674
+ "By default, audio redaction provides redacted audio URLs only when speech is detected. However, if your use-case specifically requires redacted audio files even for silent audio files without any dialogue, you can opt to receive these URLs by setting this parameter to `true`."
13675
+ ),
13676
+ override_audio_redaction_method: import_zod3.z.enum(["silence"]).optional().describe(
13677
+ "Specify the method used to redact audio. By default, redacted audio uses a beep sound. Set to `silence` to replace PII with silence instead of a beep."
13678
+ )
13679
+ }).optional().describe(
13680
+ "The options for PII-redacted audio, if redact_pii_audio is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more information.\n"
13000
13681
  ),
13001
13682
  redact_pii_audio_quality: import_zod3.z.enum(["mp3", "wav"]).describe(
13002
- "Controls the filetype of the audio created by redact_pii_audio. Currently supports mp3 (default) and wav. See [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more details."
13683
+ "Controls the filetype of the audio created by redact_pii_audio. Currently supports mp3 (default) and wav. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more details."
13003
13684
  ).or(import_zod3.z.null()).optional().describe(
13004
- "The audio quality of the PII-redacted audio file, if redact_pii_audio is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more information.\n"
13685
+ "The audio quality of the PII-redacted audio file, if redact_pii_audio is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more information.\n"
13005
13686
  ),
13006
13687
  redact_pii_policies: import_zod3.z.array(
13007
13688
  import_zod3.z.enum([
@@ -13051,19 +13732,307 @@ var getTranscriptResponse = import_zod3.z.object({
13051
13732
  "zodiac_sign"
13052
13733
  ]).describe("The type of PII to redact")
13053
13734
  ).nullish().describe(
13054
- "The list of PII Redaction policies that were enabled, if PII Redaction is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more information.\n"
13735
+ "The list of PII Redaction policies that were enabled, if PII Redaction is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/pii-redaction) for more information.\n"
13055
13736
  ),
13056
13737
  redact_pii_sub: import_zod3.z.enum(["entity_name", "hash"]).optional().describe(
13057
- 'The replacement logic for detected PII, can be "entity_name" or "hash". See [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more details.'
13738
+ "The replacement logic for detected PII, can be `entity_name` or `hash`. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction) for more details."
13739
+ ),
13740
+ sentiment_analysis: import_zod3.z.boolean().nullish().describe(
13741
+ "Whether [Sentiment Analysis](https://www.assemblyai.com/docs/speech-understanding/sentiment-analysis) is enabled, can be true or false"
13742
+ ),
13743
+ sentiment_analysis_results: import_zod3.z.array(
13744
+ import_zod3.z.object({
13745
+ text: import_zod3.z.string().describe("The transcript of the sentence"),
13746
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, of the sentence"),
13747
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, of the sentence"),
13748
+ sentiment: import_zod3.z.enum(["POSITIVE", "NEUTRAL", "NEGATIVE"]),
13749
+ confidence: import_zod3.z.number().describe(
13750
+ "The confidence score for the detected sentiment of the sentence, from 0 to 1"
13751
+ ),
13752
+ channel: import_zod3.z.string().nullish().describe(
13753
+ "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13754
+ ),
13755
+ speaker: import_zod3.z.string().nullable().describe(
13756
+ "The speaker of the sentence if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
13757
+ )
13758
+ }).describe("The result of the Sentiment Analysis model")
13759
+ ).nullish().describe(
13760
+ "An array of results for the Sentiment Analysis model, if it is enabled.\nSee [Sentiment Analysis](https://www.assemblyai.com/docs/speech-understanding/sentiment-analysis) for more information.\n"
13058
13761
  ),
13059
13762
  speaker_labels: import_zod3.z.boolean().nullish().describe(
13060
- "Whether [Speaker diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, can be true or false"
13763
+ "Whether [Speaker diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, can be true or false"
13061
13764
  ),
13062
13765
  speakers_expected: import_zod3.z.number().nullish().describe(
13063
- "Tell the speaker label model how many speakers it should attempt to identify. See [Speaker diarization](https://www.assemblyai.com/docs/models/speaker-diarization) for more details."
13766
+ "Tell the speaker label model how many speakers it should attempt to identify. See [Set number of speakers expected](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization#set-number-of-speakers-expected) for more details."
13767
+ ),
13768
+ speech_model_used: import_zod3.z.string().optional().describe(
13769
+ "The speech model to use for the transcription. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models."
13770
+ ),
13771
+ speech_models: import_zod3.z.array(
13772
+ import_zod3.z.string().describe(
13773
+ "The speech model to use for the transcription. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models."
13774
+ )
13775
+ ).nullish().describe(
13776
+ "List multiple speech models in priority order, allowing our system to automatically route your audio to the best available option. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models and routing behavior.\n"
13777
+ ),
13778
+ speech_threshold: import_zod3.z.number().nullish().describe(
13779
+ "Defaults to null. Reject audio files that contain less than this fraction of speech.\nValid values are in the range [0, 1] inclusive. See [Speech Threshold](https://www.assemblyai.com/docs/speech-threshold) for more details.\n"
13780
+ ),
13781
+ speech_understanding: import_zod3.z.object({
13782
+ request: import_zod3.z.object({
13783
+ translation: import_zod3.z.object({
13784
+ target_languages: import_zod3.z.array(import_zod3.z.string()).describe(
13785
+ 'List of target language codes (e.g., `["es", "de"]`). See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for supported languages.'
13786
+ ),
13787
+ formal: import_zod3.z.boolean().default(getTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault).describe(
13788
+ "Use formal language style. See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for more details."
13789
+ ),
13790
+ match_original_utterance: import_zod3.z.boolean().optional().describe(
13791
+ "When enabled with Speaker Labels, returns translated text in the utterances array. Each utterance will include a `translated_texts` key containing translations for each target language."
13792
+ )
13793
+ })
13794
+ }).describe(
13795
+ "Request body for [Translation](https://www.assemblyai.com/docs/speech-understanding/translation)."
13796
+ ).or(
13797
+ import_zod3.z.object({
13798
+ speaker_identification: import_zod3.z.object({
13799
+ speaker_type: import_zod3.z.enum(["role", "name"]).describe(
13800
+ "Type of speaker identification. See [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification) for details on each type."
13801
+ ),
13802
+ known_values: import_zod3.z.array(import_zod3.z.string()).optional().describe(
13803
+ 'Required if speaker_type is "role". Each value must be 35 characters or less.'
13804
+ ),
13805
+ speakers: import_zod3.z.array(
13806
+ import_zod3.z.object({
13807
+ role: import_zod3.z.string().optional().describe(
13808
+ 'The role of the speaker. Required when `speaker_type` is "role".'
13809
+ ),
13810
+ name: import_zod3.z.string().optional().describe(
13811
+ 'The name of the speaker. Required when `speaker_type` is "name".'
13812
+ ),
13813
+ description: import_zod3.z.string().optional().describe(
13814
+ "A description of the speaker to help the model identify them based on conversational context."
13815
+ )
13816
+ })
13817
+ ).optional().describe(
13818
+ "An array of speaker objects with metadata to improve identification accuracy. Each object should include a `role` or `name` (depending on `speaker_type`) and an optional `description` to help the model identify the speaker. You can also include any additional custom properties (e.g., `company`, `title`) to provide more context. Use this as an alternative to `known_values` when you want to provide additional context about each speaker."
13819
+ )
13820
+ })
13821
+ }).describe(
13822
+ "Request body for [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification)."
13823
+ )
13824
+ ).or(
13825
+ import_zod3.z.object({
13826
+ custom_formatting: import_zod3.z.object({
13827
+ date: import_zod3.z.string().optional().describe(
13828
+ 'Date format pattern (e.g., `"mm/dd/yyyy"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
13829
+ ),
13830
+ phone_number: import_zod3.z.string().optional().describe(
13831
+ 'Phone number format pattern (e.g., `"(xxx)xxx-xxxx"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
13832
+ ),
13833
+ email: import_zod3.z.string().optional().describe(
13834
+ 'Email format pattern (e.g., `"username@domain.com"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
13835
+ )
13836
+ })
13837
+ }).describe(
13838
+ "Request body for [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting)."
13839
+ )
13840
+ ).optional(),
13841
+ response: import_zod3.z.object({
13842
+ translation: import_zod3.z.object({
13843
+ status: import_zod3.z.string().optional()
13844
+ }).optional()
13845
+ }).or(
13846
+ import_zod3.z.object({
13847
+ speaker_identification: import_zod3.z.object({
13848
+ mapping: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional().describe(
13849
+ 'A mapping of the original generic speaker labels (e.g., "A", "B") to the identified speaker names or roles.'
13850
+ ),
13851
+ status: import_zod3.z.string().optional()
13852
+ }).optional()
13853
+ })
13854
+ ).or(
13855
+ import_zod3.z.object({
13856
+ custom_formatting: import_zod3.z.object({
13857
+ mapping: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional(),
13858
+ formatted_text: import_zod3.z.string().optional()
13859
+ }).optional()
13860
+ })
13861
+ ).optional()
13862
+ }).optional().describe(
13863
+ "Speech understanding tasks like [Translation](https://www.assemblyai.com/docs/speech-understanding/translation), [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification), and [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting). See the task-specific docs for available options and configuration.\n"
13864
+ ),
13865
+ status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).describe(
13866
+ "The status of your transcript. Possible values are queued, processing, completed, or error."
13867
+ ),
13868
+ summarization: import_zod3.z.boolean().describe(
13869
+ "Whether [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization) is enabled, either true or false. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details.\n\nNote: This parameter is only supported for the Universal-2 model.\n"
13870
+ ),
13871
+ summary: import_zod3.z.string().nullish().describe(
13872
+ "The generated summary of the media file, if [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization) is enabled. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details."
13873
+ ),
13874
+ summary_model: import_zod3.z.string().nullish().describe(
13875
+ "The Summarization model used to generate the summary,\nif [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization#summary-models) is enabled. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details.\n"
13876
+ ),
13877
+ summary_type: import_zod3.z.string().nullish().describe(
13878
+ "The type of summary generated, if [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization#summary-types) is enabled. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details."
13879
+ ),
13880
+ remove_audio_tags: import_zod3.z.enum(["all"]).describe(
13881
+ "Whether [audio event tags](https://www.assemblyai.com/docs/pre-recorded-audio/universal-3-pro#audio-event-tags) were removed from the transcript text.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n"
13882
+ ).or(import_zod3.z.null()).optional().describe(
13883
+ "Whether [audio event tags](https://www.assemblyai.com/docs/pre-recorded-audio/universal-3-pro#audio-event-tags) were removed from the transcript text.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n"
13884
+ ),
13885
+ temperature: import_zod3.z.number().nullish().describe(
13886
+ "The temperature that was used for the model's response. See the [Prompting Guide](https://www.assemblyai.com/docs/pre-recorded-audio/prompting) for more details.\n\nNote: This parameter can only be used with the Universal-3 Pro model.\n"
13887
+ ),
13888
+ text: import_zod3.z.string().nullish().describe("The textual transcript of your media file"),
13889
+ throttled: import_zod3.z.boolean().nullish().describe(
13890
+ "True while a request is throttled and false when a request is no longer throttled"
13891
+ ),
13892
+ utterances: import_zod3.z.array(
13893
+ import_zod3.z.object({
13894
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this utterance"),
13895
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, of the utterance in the audio file"),
13896
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, of the utterance in the audio file"),
13897
+ text: import_zod3.z.string().describe("The text for this utterance"),
13898
+ words: import_zod3.z.array(
13899
+ import_zod3.z.object({
13900
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
13901
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
13902
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
13903
+ text: import_zod3.z.string().describe("The text of the word"),
13904
+ channel: import_zod3.z.string().nullish().describe(
13905
+ "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13906
+ ),
13907
+ speaker: import_zod3.z.string().nullable().describe(
13908
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
13909
+ )
13910
+ })
13911
+ ).describe("The words in the utterance."),
13912
+ channel: import_zod3.z.string().nullish().describe(
13913
+ "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13914
+ ),
13915
+ speaker: import_zod3.z.string().describe(
13916
+ 'The speaker of this utterance, where each speaker is assigned a sequential capital letter - e.g. "A" for Speaker A, "B" for Speaker B, etc.'
13917
+ ),
13918
+ translated_texts: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional().describe(
13919
+ 'Translations keyed by language code (e.g., `{"es": "Texto traducido", "de": "\xDCbersetzter Text"}`). Only present when `match_original_utterance` is enabled with translation.'
13920
+ )
13921
+ })
13922
+ ).nullish().describe(
13923
+ "When multichannel or speaker_labels is enabled, a list of turn-by-turn utterance objects.\nSee [Speaker diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) and [Multichannel transcription](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) for more information.\n"
13924
+ ),
13925
+ webhook_auth: import_zod3.z.boolean().describe(
13926
+ "Whether [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) authentication details were provided"
13927
+ ),
13928
+ webhook_auth_header_name: import_zod3.z.string().nullish().describe(
13929
+ "The header name to be sent with the transcript completed or failed [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) requests"
13930
+ ),
13931
+ webhook_status_code: import_zod3.z.number().nullish().describe(
13932
+ "The status code we received from your server when delivering the transcript completed or failed [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) request, if a webhook URL was provided"
13933
+ ),
13934
+ webhook_url: import_zod3.z.string().nullish().describe(
13935
+ "The URL to which we send [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) requests.\n"
13936
+ ),
13937
+ words: import_zod3.z.array(
13938
+ import_zod3.z.object({
13939
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
13940
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
13941
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
13942
+ text: import_zod3.z.string().describe("The text of the word"),
13943
+ channel: import_zod3.z.string().nullish().describe(
13944
+ "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13945
+ ),
13946
+ speaker: import_zod3.z.string().nullable().describe(
13947
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
13948
+ )
13949
+ })
13950
+ ).nullish().describe(
13951
+ "An array of temporally-sequential word objects, one for each word in the transcript.\n"
13952
+ ),
13953
+ acoustic_model: import_zod3.z.string().describe("This parameter does not currently have any functionality attached to it."),
13954
+ custom_topics: import_zod3.z.boolean().nullish().describe("This parameter does not currently have any functionality attached to it."),
13955
+ language_model: import_zod3.z.string().describe("This parameter does not currently have any functionality attached to it."),
13956
+ speech_model: import_zod3.z.string().describe(
13957
+ "The speech model to use for the transcription. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models."
13958
+ ).or(import_zod3.z.null()).describe(
13959
+ "This parameter has been replaced with the `speech_models` parameter, learn more about the `speech_models` parameter [here](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model).\n"
13960
+ ),
13961
+ speed_boost: import_zod3.z.boolean().nullish().describe("This parameter does not currently have any functionality attached to it."),
13962
+ topics: import_zod3.z.array(import_zod3.z.string()).optional().describe("This parameter does not currently have any functionality attached to it."),
13963
+ translated_texts: import_zod3.z.object({
13964
+ language_code: import_zod3.z.string().optional().describe("Translated text for this language code")
13965
+ }).optional().describe(
13966
+ "Translated text keyed by language code. See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for more details."
13967
+ )
13968
+ }).describe("A transcript object");
13969
+ var deleteTranscriptParams = import_zod3.z.object({
13970
+ transcript_id: import_zod3.z.string().describe("ID of the transcript")
13971
+ });
13972
+ var deleteTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault = "auto";
13973
+ var deleteTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault = false;
13974
+ var deleteTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault = 0.3;
13975
+ var deleteTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault = false;
13976
+ var deleteTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault = true;
13977
+ var deleteTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault = false;
13978
+ var deleteTranscriptResponse = import_zod3.z.object({
13979
+ audio_channels: import_zod3.z.number().optional().describe(
13980
+ "The number of audio channels in the audio file. This is only present when [multichannel](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) is enabled."
13981
+ ),
13982
+ audio_duration: import_zod3.z.number().nullish().describe("The duration of this transcript object's media file, in seconds"),
13983
+ audio_end_at: import_zod3.z.number().nullish().describe(
13984
+ "The point in time, in milliseconds, in the file at which the transcription was terminated. See [Set the start and end of the transcript](https://www.assemblyai.com/docs/pre-recorded-audio/set-the-start-and-end-of-the-transcript) for more details."
13985
+ ),
13986
+ audio_start_from: import_zod3.z.number().nullish().describe(
13987
+ "The point in time, in milliseconds, in the file at which the transcription was started. See [Set the start and end of the transcript](https://www.assemblyai.com/docs/pre-recorded-audio/set-the-start-and-end-of-the-transcript) for more details."
13988
+ ),
13989
+ audio_url: import_zod3.z.string().describe("The URL of the media that was transcribed"),
13990
+ auto_chapters: import_zod3.z.boolean().nullish().describe(
13991
+ "Whether [Auto Chapters](https://www.assemblyai.com/docs/speech-understanding/auto-chapters) is enabled, can be true or false. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible chapter summaries. See the [updated Auto Chapters page](https://www.assemblyai.com/docs/speech-understanding/auto-chapters) for details.\n\nNote: This parameter is only supported for the Universal-2 model.\n"
13992
+ ),
13993
+ auto_highlights: import_zod3.z.boolean().describe(
13994
+ "Whether [Key Phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases) is enabled, either true or false"
13995
+ ),
13996
+ auto_highlights_result: import_zod3.z.object({
13997
+ status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
13998
+ results: import_zod3.z.array(
13999
+ import_zod3.z.object({
14000
+ count: import_zod3.z.number().describe("The total number of times the key phrase appears in the audio file"),
14001
+ rank: import_zod3.z.number().describe(
14002
+ "The total relevancy to the overall audio file of this key phrase - a greater number means more relevant"
14003
+ ),
14004
+ text: import_zod3.z.string().describe("The text itself of the key phrase"),
14005
+ timestamps: import_zod3.z.array(
14006
+ import_zod3.z.object({
14007
+ start: import_zod3.z.number().describe("The start time in milliseconds"),
14008
+ end: import_zod3.z.number().describe("The end time in milliseconds")
14009
+ }).describe("Timestamp containing a start and end property in milliseconds")
14010
+ ).describe("The timestamp of the of the key phrase")
14011
+ })
14012
+ ).describe("A temporally-sequential array of Key Phrases")
14013
+ }).describe(
14014
+ "An array of results for the Key Phrases model, if it is enabled.\nSee [Key phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases) for more information.\n"
14015
+ ).or(import_zod3.z.null()).optional().describe(
14016
+ "An array of results for the Key Phrases model, if it is enabled.\nSee [Key Phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases) for more information.\n"
14017
+ ),
14018
+ chapters: import_zod3.z.array(
14019
+ import_zod3.z.object({
14020
+ gist: import_zod3.z.string().describe(
14021
+ "An ultra-short summary (just a few words) of the content spoken in the chapter"
14022
+ ),
14023
+ headline: import_zod3.z.string().describe("A single sentence summary of the content spoken during the chapter"),
14024
+ summary: import_zod3.z.string().describe("A one paragraph summary of the content spoken during the chapter"),
14025
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter"),
14026
+ end: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter")
14027
+ }).describe("Chapter of the audio file")
14028
+ ).nullish().describe(
14029
+ "An array of temporally sequential chapters for the audio file. See [Auto Chapters](https://www.assemblyai.com/docs/speech-understanding/auto-chapters) for more information."
14030
+ ),
14031
+ confidence: import_zod3.z.number().nullish().describe(
14032
+ "The confidence score for the transcript, between 0.0 (low confidence) and 1.0 (high confidence)"
13064
14033
  ),
13065
14034
  content_safety: import_zod3.z.boolean().nullish().describe(
13066
- "Whether [Content Moderation](https://www.assemblyai.com/docs/models/content-moderation) is enabled, can be true or false"
14035
+ "Whether [Content Moderation](https://www.assemblyai.com/docs/content-moderation) is enabled, can be true or false"
13067
14036
  ),
13068
14037
  content_safety_labels: import_zod3.z.object({
13069
14038
  status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
@@ -13073,12 +14042,8 @@ var getTranscriptResponse = import_zod3.z.object({
13073
14042
  labels: import_zod3.z.array(
13074
14043
  import_zod3.z.object({
13075
14044
  label: import_zod3.z.string().describe("The label of the sensitive topic"),
13076
- confidence: import_zod3.z.number().min(
13077
- getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin
13078
- ).max(
13079
- getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax
13080
- ).describe("The confidence score for the topic being discussed, from 0 to 1"),
13081
- severity: import_zod3.z.number().min(getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin).max(getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax).describe("How severely the topic is discussed in the section, from 0 to 1")
14045
+ confidence: import_zod3.z.number().describe("The confidence score for the topic being discussed, from 0 to 1"),
14046
+ severity: import_zod3.z.number().describe("How severely the topic is discussed in the section, from 0 to 1")
13082
14047
  })
13083
14048
  ).describe(
13084
14049
  "An array of safety labels, one per sensitive topic that was detected in the section"
@@ -13091,128 +14056,39 @@ var getTranscriptResponse = import_zod3.z.object({
13091
14056
  }).describe("Timestamp containing a start and end property in milliseconds")
13092
14057
  })
13093
14058
  ).describe("An array of results for the Content Moderation model"),
13094
- summary: import_zod3.z.record(
13095
- import_zod3.z.string(),
13096
- import_zod3.z.number().min(getTranscriptResponseContentSafetyLabelsSummaryMinOne).max(getTranscriptResponseContentSafetyLabelsSummaryMaxOne).describe(
13097
- 'A confidence score for the presence of the sensitive topic "topic" across the entire audio file'
13098
- )
13099
- ).describe(
14059
+ summary: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.number()).describe(
13100
14060
  "A summary of the Content Moderation confidence results for the entire audio file"
13101
14061
  ),
13102
14062
  severity_score_summary: import_zod3.z.record(
13103
14063
  import_zod3.z.string(),
13104
14064
  import_zod3.z.object({
13105
- low: import_zod3.z.number().min(getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin).max(getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax),
13106
- medium: import_zod3.z.number().min(getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin).max(getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax),
13107
- high: import_zod3.z.number().min(getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin).max(getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax)
14065
+ low: import_zod3.z.number(),
14066
+ medium: import_zod3.z.number(),
14067
+ high: import_zod3.z.number()
13108
14068
  })
13109
14069
  ).describe(
13110
14070
  "A summary of the Content Moderation severity results for the entire audio file"
13111
14071
  )
13112
14072
  }).describe(
13113
- "An array of results for the Content Moderation model, if it is enabled.\nSee [Content moderation](https://www.assemblyai.com/docs/models/content-moderation) for more information.\n"
13114
- ).or(import_zod3.z.null()).optional().describe(
13115
- "An array of results for the Content Moderation model, if it is enabled.\nSee [Content moderation](https://www.assemblyai.com/docs/models/content-moderation) for more information.\n"
13116
- ),
13117
- iab_categories: import_zod3.z.boolean().nullish().describe(
13118
- "Whether [Topic Detection](https://www.assemblyai.com/docs/models/topic-detection) is enabled, can be true or false"
13119
- ),
13120
- iab_categories_result: import_zod3.z.object({
13121
- status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
13122
- results: import_zod3.z.array(
13123
- import_zod3.z.object({
13124
- text: import_zod3.z.string().describe("The text in the transcript in which a detected topic occurs"),
13125
- labels: import_zod3.z.array(
13126
- import_zod3.z.object({
13127
- relevance: import_zod3.z.number().min(
13128
- getTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin
13129
- ).max(
13130
- getTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax
13131
- ).describe("How relevant the detected topic is of a detected topic"),
13132
- label: import_zod3.z.string().describe(
13133
- "The IAB taxonomical label for the label of the detected topic, where > denotes supertopic/subtopic relationship"
13134
- )
13135
- })
13136
- ).optional().describe("An array of detected topics in the text"),
13137
- timestamp: import_zod3.z.object({
13138
- start: import_zod3.z.number().describe("The start time in milliseconds"),
13139
- end: import_zod3.z.number().describe("The end time in milliseconds")
13140
- }).optional().describe("Timestamp containing a start and end property in milliseconds")
13141
- }).describe("The result of the topic detection model")
13142
- ).describe("An array of results for the Topic Detection model"),
13143
- summary: import_zod3.z.record(
13144
- import_zod3.z.string(),
13145
- import_zod3.z.number().min(getTranscriptResponseIabCategoriesResultSummaryMinOne).max(getTranscriptResponseIabCategoriesResultSummaryMaxOne)
13146
- ).describe("The overall relevance of topic to the entire audio file")
13147
- }).describe(
13148
- "The result of the Topic Detection model, if it is enabled.\nSee [Topic Detection](https://www.assemblyai.com/docs/models/topic-detection) for more information.\n"
14073
+ "An array of results for the Content Moderation model, if it is enabled.\nSee [Content moderation](https://www.assemblyai.com/docs/content-moderation) for more information.\n"
13149
14074
  ).or(import_zod3.z.null()).optional().describe(
13150
- "The result of the Topic Detection model, if it is enabled.\nSee [Topic Detection](https://www.assemblyai.com/docs/models/topic-detection) for more information.\n"
14075
+ "An array of results for the Content Moderation model, if it is enabled.\nSee [Content moderation](https://www.assemblyai.com/docs/content-moderation) for more information.\n"
13151
14076
  ),
13152
14077
  custom_spelling: import_zod3.z.array(
13153
14078
  import_zod3.z.object({
13154
- from: import_zod3.z.array(import_zod3.z.string().describe("Word or phrase to replace")).describe("Words or phrases to replace"),
14079
+ from: import_zod3.z.array(import_zod3.z.string()).describe("Words or phrases to replace"),
13155
14080
  to: import_zod3.z.string().describe("Word to replace with")
13156
14081
  }).describe(
13157
14082
  "Object containing words or phrases to replace, and the word or phrase to replace with"
13158
14083
  )
13159
- ).nullish().describe("Customize how words are spelled and formatted using to and from values"),
13160
- keyterms_prompt: import_zod3.z.array(import_zod3.z.string()).optional().describe(
13161
- "Improve accuracy with up to 1000 domain-specific words or phrases (maximum 6 words per phrase).\n"
13162
- ),
13163
- prompt: import_zod3.z.string().optional().describe("This parameter does not currently have any functionality attached to it."),
13164
- auto_chapters: import_zod3.z.boolean().nullish().describe(
13165
- "Whether [Auto Chapters](https://www.assemblyai.com/docs/models/auto-chapters) is enabled, can be true or false"
13166
- ),
13167
- chapters: import_zod3.z.array(
13168
- import_zod3.z.object({
13169
- gist: import_zod3.z.string().describe(
13170
- "An ultra-short summary (just a few words) of the content spoken in the chapter"
13171
- ),
13172
- headline: import_zod3.z.string().describe("A single sentence summary of the content spoken during the chapter"),
13173
- summary: import_zod3.z.string().describe("A one paragraph summary of the content spoken during the chapter"),
13174
- start: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter"),
13175
- end: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter")
13176
- }).describe("Chapter of the audio file")
13177
- ).nullish().describe("An array of temporally sequential chapters for the audio file"),
13178
- summarization: import_zod3.z.boolean().describe(
13179
- "Whether [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled, either true or false"
13180
- ),
13181
- summary_type: import_zod3.z.string().nullish().describe(
13182
- "The type of summary generated, if [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled"
13183
- ),
13184
- summary_model: import_zod3.z.string().nullish().describe(
13185
- "The Summarization model used to generate the summary,\nif [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled\n"
13186
- ),
13187
- summary: import_zod3.z.string().nullish().describe(
13188
- "The generated summary of the media file, if [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled"
13189
- ),
13190
- custom_topics: import_zod3.z.boolean().nullish().describe("Whether custom topics is enabled, either true or false"),
13191
- topics: import_zod3.z.array(import_zod3.z.string()).optional().describe("The list of custom topics provided if custom topics is enabled"),
13192
- sentiment_analysis: import_zod3.z.boolean().nullish().describe(
13193
- "Whether [Sentiment Analysis](https://www.assemblyai.com/docs/models/sentiment-analysis) is enabled, can be true or false"
13194
- ),
13195
- sentiment_analysis_results: import_zod3.z.array(
13196
- import_zod3.z.object({
13197
- text: import_zod3.z.string().describe("The transcript of the sentence"),
13198
- start: import_zod3.z.number().describe("The starting time, in milliseconds, of the sentence"),
13199
- end: import_zod3.z.number().describe("The ending time, in milliseconds, of the sentence"),
13200
- sentiment: import_zod3.z.enum(["POSITIVE", "NEUTRAL", "NEGATIVE"]),
13201
- confidence: import_zod3.z.number().min(getTranscriptResponseSentimentAnalysisResultsItemConfidenceMin).max(getTranscriptResponseSentimentAnalysisResultsItemConfidenceMax).describe(
13202
- "The confidence score for the detected sentiment of the sentence, from 0 to 1"
13203
- ),
13204
- channel: import_zod3.z.string().nullish().describe(
13205
- "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13206
- ),
13207
- speaker: import_zod3.z.string().nullable().describe(
13208
- "The speaker of the sentence if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
13209
- )
13210
- }).describe("The result of the Sentiment Analysis model")
13211
14084
  ).nullish().describe(
13212
- "An array of results for the Sentiment Analysis model, if it is enabled.\nSee [Sentiment Analysis](https://www.assemblyai.com/docs/models/sentiment-analysis) for more information.\n"
14085
+ "Customize how words are spelled and formatted using to and from values. See [Custom Spelling](https://www.assemblyai.com/docs/pre-recorded-audio/custom-spelling) for more details."
13213
14086
  ),
13214
- entity_detection: import_zod3.z.boolean().nullish().describe(
13215
- "Whether [Entity Detection](https://www.assemblyai.com/docs/models/entity-detection) is enabled, can be true or false"
14087
+ disfluencies: import_zod3.z.boolean().nullish().describe(
14088
+ 'Transcribe [Filler Words](https://www.assemblyai.com/docs/pre-recorded-audio/filler-words), like "umm", in your media file; can be true or false'
14089
+ ),
14090
+ domain: import_zod3.z.string().nullish().describe(
14091
+ 'The domain-specific model applied to the transcript. When set to `"medical-v1"`, [Medical Mode](https://www.assemblyai.com/docs/pre-recorded-audio/medical-mode) was used to improve accuracy for medical terminology.\n'
13216
14092
  ),
13217
14093
  entities: import_zod3.z.array(
13218
14094
  import_zod3.z.object({
@@ -13271,61 +14147,49 @@ var getTranscriptResponse = import_zod3.z.object({
13271
14147
  )
13272
14148
  }).describe("A detected entity")
13273
14149
  ).nullish().describe(
13274
- "An array of results for the Entity Detection model, if it is enabled.\nSee [Entity detection](https://www.assemblyai.com/docs/models/entity-detection) for more information.\n"
14150
+ "An array of results for the Entity Detection model, if it is enabled.\nSee [Entity detection](https://www.assemblyai.com/docs/speech-understanding/entity-detection) for more information.\n"
13275
14151
  ),
13276
- speech_threshold: import_zod3.z.number().min(getTranscriptResponseSpeechThresholdMin).max(getTranscriptResponseSpeechThresholdMax).nullish().describe(
13277
- "Defaults to null. Reject audio files that contain less than this fraction of speech.\nValid values are in the range [0, 1] inclusive.\n"
13278
- ),
13279
- throttled: import_zod3.z.boolean().nullish().describe(
13280
- "True while a request is throttled and false when a request is no longer throttled"
14152
+ entity_detection: import_zod3.z.boolean().nullish().describe(
14153
+ "Whether [Entity Detection](https://www.assemblyai.com/docs/speech-understanding/entity-detection) is enabled, can be true or false"
13281
14154
  ),
13282
14155
  error: import_zod3.z.string().optional().describe("Error message of why the transcript failed"),
13283
- language_model: import_zod3.z.string().describe("The language model that was used for the transcript"),
13284
- acoustic_model: import_zod3.z.string().describe("The acoustic model that was used for the transcript")
13285
- }).describe("A transcript object");
13286
- var deleteTranscriptParams = import_zod3.z.object({
13287
- transcript_id: import_zod3.z.string().describe("ID of the transcript")
13288
- });
13289
- var deleteTranscriptResponseLanguageConfidenceThresholdMin = 0;
13290
- var deleteTranscriptResponseLanguageConfidenceThresholdMax = 1;
13291
- var deleteTranscriptResponseLanguageConfidenceMin = 0;
13292
- var deleteTranscriptResponseLanguageConfidenceMax = 1;
13293
- var deleteTranscriptResponseSpeechModelDefault = null;
13294
- var deleteTranscriptResponseWordsItemConfidenceMin = 0;
13295
- var deleteTranscriptResponseWordsItemConfidenceMax = 1;
13296
- var deleteTranscriptResponseUtterancesItemConfidenceMin = 0;
13297
- var deleteTranscriptResponseUtterancesItemConfidenceMax = 1;
13298
- var deleteTranscriptResponseUtterancesItemWordsItemConfidenceMin = 0;
13299
- var deleteTranscriptResponseUtterancesItemWordsItemConfidenceMax = 1;
13300
- var deleteTranscriptResponseConfidenceMin = 0;
13301
- var deleteTranscriptResponseConfidenceMax = 1;
13302
- var deleteTranscriptResponseAutoHighlightsResultResultsItemRankMin = 0;
13303
- var deleteTranscriptResponseAutoHighlightsResultResultsItemRankMax = 1;
13304
- var deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin = 0;
13305
- var deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax = 1;
13306
- var deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin = 0;
13307
- var deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax = 1;
13308
- var deleteTranscriptResponseContentSafetyLabelsSummaryMinOne = 0;
13309
- var deleteTranscriptResponseContentSafetyLabelsSummaryMaxOne = 1;
13310
- var deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin = 0;
13311
- var deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax = 1;
13312
- var deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin = 0;
13313
- var deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax = 1;
13314
- var deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin = 0;
13315
- var deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax = 1;
13316
- var deleteTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin = 0;
13317
- var deleteTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax = 1;
13318
- var deleteTranscriptResponseIabCategoriesResultSummaryMinOne = 0;
13319
- var deleteTranscriptResponseIabCategoriesResultSummaryMaxOne = 1;
13320
- var deleteTranscriptResponseSentimentAnalysisResultsItemConfidenceMin = 0;
13321
- var deleteTranscriptResponseSentimentAnalysisResultsItemConfidenceMax = 1;
13322
- var deleteTranscriptResponseSpeechThresholdMin = 0;
13323
- var deleteTranscriptResponseSpeechThresholdMax = 1;
13324
- var deleteTranscriptResponse = import_zod3.z.object({
14156
+ filter_profanity: import_zod3.z.boolean().nullish().describe(
14157
+ "Whether [Profanity Filtering](https://www.assemblyai.com/docs/profanity-filtering) is enabled, either true or false"
14158
+ ),
14159
+ format_text: import_zod3.z.boolean().nullish().describe(
14160
+ "Whether [Text Formatting](https://www.assemblyai.com/docs/pre-recorded-audio) is enabled, either true or false"
14161
+ ),
14162
+ iab_categories: import_zod3.z.boolean().nullish().describe(
14163
+ "Whether [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection) is enabled, can be true or false"
14164
+ ),
14165
+ iab_categories_result: import_zod3.z.object({
14166
+ status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
14167
+ results: import_zod3.z.array(
14168
+ import_zod3.z.object({
14169
+ text: import_zod3.z.string().describe("The text in the transcript in which a detected topic occurs"),
14170
+ labels: import_zod3.z.array(
14171
+ import_zod3.z.object({
14172
+ relevance: import_zod3.z.number().describe("How relevant the detected topic is of a detected topic"),
14173
+ label: import_zod3.z.string().describe(
14174
+ "The IAB taxonomical label for the label of the detected topic, where > denotes supertopic/subtopic relationship"
14175
+ )
14176
+ })
14177
+ ).optional().describe("An array of detected topics in the text"),
14178
+ timestamp: import_zod3.z.object({
14179
+ start: import_zod3.z.number().describe("The start time in milliseconds"),
14180
+ end: import_zod3.z.number().describe("The end time in milliseconds")
14181
+ }).optional().describe("Timestamp containing a start and end property in milliseconds")
14182
+ }).describe("The result of the topic detection model")
14183
+ ).describe("An array of results for the Topic Detection model"),
14184
+ summary: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.number()).describe("The overall relevance of topic to the entire audio file")
14185
+ }).describe(
14186
+ "The result of the Topic Detection model, if it is enabled.\nSee [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection) for more information.\n"
14187
+ ).or(import_zod3.z.null()).optional().describe(
14188
+ "The result of the Topic Detection model, if it is enabled.\nSee [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection) for more information.\n"
14189
+ ),
13325
14190
  id: import_zod3.z.string().uuid().describe("The unique identifier of your transcript"),
13326
- audio_url: import_zod3.z.string().describe("The URL of the media that was transcribed"),
13327
- status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).describe(
13328
- "The status of your transcript. Possible values are queued, processing, completed, or error."
14191
+ keyterms_prompt: import_zod3.z.array(import_zod3.z.string()).optional().describe(
14192
+ "Improve accuracy with up to 200 (for Universal-2) or 1000 (for Universal-3 Pro) domain-specific words or phrases (maximum 6 words per phrase). See [Keyterms Prompting](https://www.assemblyai.com/docs/pre-recorded-audio/keyterms-prompting) for more details.\n"
13329
14193
  ),
13330
14194
  language_code: import_zod3.z.enum([
13331
14195
  "en",
@@ -13430,138 +14294,175 @@ var deleteTranscriptResponse = import_zod3.z.object({
13430
14294
  "cy",
13431
14295
  "yi",
13432
14296
  "yo"
13433
- ]).describe(
13434
- "The language of your audio file. Possible values are found in [Supported Languages](https://www.assemblyai.com/docs/concepts/supported-languages).\nThe default value is 'en_us'.\n"
13435
- ).or(import_zod3.z.string()).optional().describe(
13436
- "The language of your audio file.\nPossible values are found in [Supported Languages](https://www.assemblyai.com/docs/concepts/supported-languages).\nThe default value is 'en_us'.\n"
13437
- ),
13438
- language_detection: import_zod3.z.boolean().nullish().describe(
13439
- "Whether [Automatic language detection](https://www.assemblyai.com/docs/models/speech-recognition#automatic-language-detection) is enabled, either true or false"
13440
- ),
13441
- language_confidence_threshold: import_zod3.z.number().min(deleteTranscriptResponseLanguageConfidenceThresholdMin).max(deleteTranscriptResponseLanguageConfidenceThresholdMax).nullable().describe(
13442
- "The confidence threshold for the automatically detected language.\nAn error will be returned if the language confidence is below this threshold.\n"
13443
- ),
13444
- language_confidence: import_zod3.z.number().min(deleteTranscriptResponseLanguageConfidenceMin).max(deleteTranscriptResponseLanguageConfidenceMax).nullable().describe(
13445
- "The confidence score for the detected language, between 0.0 (low confidence) and 1.0 (high confidence)"
13446
- ),
13447
- speech_model: import_zod3.z.enum(["best", "slam-1", "universal"]).describe("The speech model to use for the transcription.").or(import_zod3.z.null()).describe(
13448
- "The speech model used for the transcription. When `null`, the default model is used."
13449
- ),
13450
- text: import_zod3.z.string().nullish().describe("The textual transcript of your media file"),
13451
- words: import_zod3.z.array(
13452
- import_zod3.z.object({
13453
- confidence: import_zod3.z.number().min(deleteTranscriptResponseWordsItemConfidenceMin).max(deleteTranscriptResponseWordsItemConfidenceMax).describe("The confidence score for the transcript of this word"),
13454
- start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
13455
- end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
13456
- text: import_zod3.z.string().describe("The text of the word"),
13457
- channel: import_zod3.z.string().nullish().describe(
13458
- "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13459
- ),
13460
- speaker: import_zod3.z.string().nullable().describe(
13461
- "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
13462
- )
13463
- })
13464
- ).nullish().describe(
13465
- "An array of temporally-sequential word objects, one for each word in the transcript.\nSee [Speech recognition](https://www.assemblyai.com/docs/models/speech-recognition) for more information.\n"
14297
+ ]).optional().describe(
14298
+ "The language of your audio file. Possible values are found in [Supported Languages](https://www.assemblyai.com/docs/pre-recorded-audio/supported-languages).\nThe default value is 'en_us'.\n"
13466
14299
  ),
13467
- utterances: import_zod3.z.array(
13468
- import_zod3.z.object({
13469
- confidence: import_zod3.z.number().min(deleteTranscriptResponseUtterancesItemConfidenceMin).max(deleteTranscriptResponseUtterancesItemConfidenceMax).describe("The confidence score for the transcript of this utterance"),
13470
- start: import_zod3.z.number().describe("The starting time, in milliseconds, of the utterance in the audio file"),
13471
- end: import_zod3.z.number().describe("The ending time, in milliseconds, of the utterance in the audio file"),
13472
- text: import_zod3.z.string().describe("The text for this utterance"),
13473
- words: import_zod3.z.array(
13474
- import_zod3.z.object({
13475
- confidence: import_zod3.z.number().min(deleteTranscriptResponseUtterancesItemWordsItemConfidenceMin).max(deleteTranscriptResponseUtterancesItemWordsItemConfidenceMax).describe("The confidence score for the transcript of this word"),
13476
- start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
13477
- end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
13478
- text: import_zod3.z.string().describe("The text of the word"),
13479
- channel: import_zod3.z.string().nullish().describe(
13480
- "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13481
- ),
13482
- speaker: import_zod3.z.string().nullable().describe(
13483
- "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
13484
- )
13485
- })
13486
- ).describe("The words in the utterance."),
13487
- channel: import_zod3.z.string().nullish().describe(
13488
- "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13489
- ),
13490
- speaker: import_zod3.z.string().describe(
13491
- 'The speaker of this utterance, where each speaker is assigned a sequential capital letter - e.g. "A" for Speaker A, "B" for Speaker B, etc.'
13492
- )
13493
- })
14300
+ language_codes: import_zod3.z.array(
14301
+ import_zod3.z.enum([
14302
+ "en",
14303
+ "en_au",
14304
+ "en_uk",
14305
+ "en_us",
14306
+ "es",
14307
+ "fr",
14308
+ "de",
14309
+ "it",
14310
+ "pt",
14311
+ "nl",
14312
+ "af",
14313
+ "sq",
14314
+ "am",
14315
+ "ar",
14316
+ "hy",
14317
+ "as",
14318
+ "az",
14319
+ "ba",
14320
+ "eu",
14321
+ "be",
14322
+ "bn",
14323
+ "bs",
14324
+ "br",
14325
+ "bg",
14326
+ "my",
14327
+ "ca",
14328
+ "zh",
14329
+ "hr",
14330
+ "cs",
14331
+ "da",
14332
+ "et",
14333
+ "fo",
14334
+ "fi",
14335
+ "gl",
14336
+ "ka",
14337
+ "el",
14338
+ "gu",
14339
+ "ht",
14340
+ "ha",
14341
+ "haw",
14342
+ "he",
14343
+ "hi",
14344
+ "hu",
14345
+ "is",
14346
+ "id",
14347
+ "ja",
14348
+ "jw",
14349
+ "kn",
14350
+ "kk",
14351
+ "km",
14352
+ "ko",
14353
+ "lo",
14354
+ "la",
14355
+ "lv",
14356
+ "ln",
14357
+ "lt",
14358
+ "lb",
14359
+ "mk",
14360
+ "mg",
14361
+ "ms",
14362
+ "ml",
14363
+ "mt",
14364
+ "mi",
14365
+ "mr",
14366
+ "mn",
14367
+ "ne",
14368
+ "no",
14369
+ "nn",
14370
+ "oc",
14371
+ "pa",
14372
+ "ps",
14373
+ "fa",
14374
+ "pl",
14375
+ "ro",
14376
+ "ru",
14377
+ "sa",
14378
+ "sr",
14379
+ "sn",
14380
+ "sd",
14381
+ "si",
14382
+ "sk",
14383
+ "sl",
14384
+ "so",
14385
+ "su",
14386
+ "sw",
14387
+ "sv",
14388
+ "tl",
14389
+ "tg",
14390
+ "ta",
14391
+ "tt",
14392
+ "te",
14393
+ "th",
14394
+ "bo",
14395
+ "tr",
14396
+ "tk",
14397
+ "uk",
14398
+ "ur",
14399
+ "uz",
14400
+ "vi",
14401
+ "cy",
14402
+ "yi",
14403
+ "yo"
14404
+ ]).describe(
14405
+ "The language of your audio file. Possible values are found in [Supported Languages](https://www.assemblyai.com/docs/pre-recorded-audio/supported-languages).\nThe default value is 'en_us'.\n"
14406
+ )
13494
14407
  ).nullish().describe(
13495
- "When multichannel or speaker_labels is enabled, a list of turn-by-turn utterance objects.\nSee [Speaker diarization](https://www.assemblyai.com/docs/speech-to-text/speaker-diarization) and [Multichannel transcription](https://www.assemblyai.com/docs/speech-to-text/speech-recognition#multichannel-transcription) for more information.\n"
13496
- ),
13497
- confidence: import_zod3.z.number().min(deleteTranscriptResponseConfidenceMin).max(deleteTranscriptResponseConfidenceMax).nullish().describe(
13498
- "The confidence score for the transcript, between 0.0 (low confidence) and 1.0 (high confidence)"
13499
- ),
13500
- audio_duration: import_zod3.z.number().nullish().describe("The duration of this transcript object's media file, in seconds"),
13501
- punctuate: import_zod3.z.boolean().nullish().describe("Whether Automatic Punctuation is enabled, either true or false"),
13502
- format_text: import_zod3.z.boolean().nullish().describe("Whether Text Formatting is enabled, either true or false"),
13503
- disfluencies: import_zod3.z.boolean().nullish().describe('Transcribe Filler Words, like "umm", in your media file; can be true or false'),
13504
- multichannel: import_zod3.z.boolean().nullish().describe(
13505
- "Whether [Multichannel transcription](https://www.assemblyai.com/docs/models/speech-recognition#multichannel-transcription) was enabled in the transcription request, either true or false"
13506
- ),
13507
- audio_channels: import_zod3.z.number().optional().describe(
13508
- "The number of audio channels in the audio file. This is only present when multichannel is enabled."
14408
+ "The language codes of your audio file. Used for [Code switching](/docs/speech-to-text/pre-recorded-audio/code-switching)\nOne of the values specified must be `en`.\n"
13509
14409
  ),
13510
- webhook_url: import_zod3.z.string().nullish().describe(
13511
- "The URL to which we send webhook requests.\nWe sends two different types of webhook requests.\nOne request when a transcript is completed or failed, and one request when the redacted audio is ready if redact_pii_audio is enabled.\n"
14410
+ language_confidence: import_zod3.z.number().nullable().describe(
14411
+ "The confidence score for the detected language, between 0.0 (low confidence) and 1.0 (high confidence). See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details."
13512
14412
  ),
13513
- webhook_status_code: import_zod3.z.number().nullish().describe(
13514
- "The status code we received from your server when delivering the transcript completed or failed webhook request, if a webhook URL was provided"
14413
+ language_confidence_threshold: import_zod3.z.number().nullable().describe(
14414
+ "The confidence threshold for the automatically detected language.\nAn error will be returned if the language confidence is below this threshold.\nSee [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.\n"
13515
14415
  ),
13516
- webhook_auth: import_zod3.z.boolean().describe("Whether webhook authentication details were provided"),
13517
- webhook_auth_header_name: import_zod3.z.string().nullish().describe(
13518
- "The header name to be sent with the transcript completed or failed webhook requests"
14416
+ language_detection: import_zod3.z.boolean().nullish().describe(
14417
+ "Whether [Automatic language detection](/docs/pre-recorded-audio/automatic-language-detection) is enabled, either true or false"
13519
14418
  ),
13520
- speed_boost: import_zod3.z.boolean().nullish().describe("Whether speed boost is enabled"),
13521
- auto_highlights: import_zod3.z.boolean().describe("Whether Key Phrases is enabled, either true or false"),
13522
- auto_highlights_result: import_zod3.z.object({
13523
- status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
13524
- results: import_zod3.z.array(
13525
- import_zod3.z.object({
13526
- count: import_zod3.z.number().describe("The total number of times the key phrase appears in the audio file"),
13527
- rank: import_zod3.z.number().min(deleteTranscriptResponseAutoHighlightsResultResultsItemRankMin).max(deleteTranscriptResponseAutoHighlightsResultResultsItemRankMax).describe(
13528
- "The total relevancy to the overall audio file of this key phrase - a greater number means more relevant"
13529
- ),
13530
- text: import_zod3.z.string().describe("The text itself of the key phrase"),
13531
- timestamps: import_zod3.z.array(
13532
- import_zod3.z.object({
13533
- start: import_zod3.z.number().describe("The start time in milliseconds"),
13534
- end: import_zod3.z.number().describe("The end time in milliseconds")
13535
- }).describe("Timestamp containing a start and end property in milliseconds")
13536
- ).describe("The timestamp of the of the key phrase")
13537
- })
13538
- ).describe("A temporally-sequential array of Key Phrases")
13539
- }).describe(
13540
- "An array of results for the Key Phrases model, if it is enabled.\nSee [Key phrases](https://www.assemblyai.com/docs/models/key-phrases) for more information.\n"
13541
- ).or(import_zod3.z.null()).optional().describe(
13542
- "An array of results for the Key Phrases model, if it is enabled.\nSee [Key Phrases](https://www.assemblyai.com/docs/models/key-phrases) for more information.\n"
14419
+ language_detection_options: import_zod3.z.object({
14420
+ expected_languages: import_zod3.z.array(import_zod3.z.string()).optional().describe(
14421
+ 'List of languages expected in the audio file. Defaults to `["all"]` when unspecified. See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.'
14422
+ ),
14423
+ fallback_language: import_zod3.z.string().default(deleteTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault).describe(
14424
+ 'If the detected language of the audio file is not in the list of expected languages, the `fallback_language` is used. Specify `["auto"]` to let our model choose the fallback language from `expected_languages` with the highest confidence score. See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.\n'
14425
+ ),
14426
+ code_switching: import_zod3.z.boolean().optional().describe(
14427
+ "Whether [code switching](/docs/speech-to-text/pre-recorded-audio/code-switching) should be detected.\n"
14428
+ ),
14429
+ code_switching_confidence_threshold: import_zod3.z.number().default(
14430
+ deleteTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault
14431
+ ).describe(
14432
+ "The confidence threshold for [code switching](/docs/speech-to-text/pre-recorded-audio/code-switching) detection. If the code switching confidence is below this threshold, the transcript will be processed in the language with the highest `language_detection_confidence` score.\n"
14433
+ )
14434
+ }).optional().describe(
14435
+ "Specify options for [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection)."
13543
14436
  ),
13544
- audio_start_from: import_zod3.z.number().nullish().describe(
13545
- "The point in time, in milliseconds, in the file at which the transcription was started"
14437
+ multichannel: import_zod3.z.boolean().nullish().describe(
14438
+ "Whether [Multichannel transcription](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) was enabled in the transcription request, either true or false"
13546
14439
  ),
13547
- audio_end_at: import_zod3.z.number().nullish().describe(
13548
- "The point in time, in milliseconds, in the file at which the transcription was terminated"
14440
+ prompt: import_zod3.z.string().optional().describe(
14441
+ "Provide natural language prompting of up to 1,500 words of contextual information to the model. See the [Prompting Guide](https://www.assemblyai.com/docs/pre-recorded-audio/prompting) for best practices.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n"
13549
14442
  ),
13550
- word_boost: import_zod3.z.array(import_zod3.z.string()).optional().describe("The list of custom vocabulary to boost transcription probability for"),
13551
- boost_param: import_zod3.z.string().nullish().describe("The word boost parameter value"),
13552
- filter_profanity: import_zod3.z.boolean().nullish().describe(
13553
- "Whether [Profanity Filtering](https://www.assemblyai.com/docs/models/speech-recognition#profanity-filtering) is enabled, either true or false"
14443
+ punctuate: import_zod3.z.boolean().nullish().describe(
14444
+ "Whether [Automatic Punctuation](https://www.assemblyai.com/docs/pre-recorded-audio) is enabled, either true or false"
13554
14445
  ),
13555
14446
  redact_pii: import_zod3.z.boolean().describe(
13556
- "Whether [PII Redaction](https://www.assemblyai.com/docs/models/pii-redaction) is enabled, either true or false"
14447
+ "Whether [PII Redaction](https://www.assemblyai.com/docs/pii-redaction) is enabled, either true or false"
13557
14448
  ),
13558
14449
  redact_pii_audio: import_zod3.z.boolean().nullish().describe(
13559
- "Whether a redacted version of the audio file was generated,\neither true or false. See [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more information.\n"
14450
+ "Whether a redacted version of the audio file was generated,\neither true or false. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more information.\n"
14451
+ ),
14452
+ redact_pii_audio_options: import_zod3.z.object({
14453
+ return_redacted_no_speech_audio: import_zod3.z.boolean().optional().describe(
14454
+ "By default, audio redaction provides redacted audio URLs only when speech is detected. However, if your use-case specifically requires redacted audio files even for silent audio files without any dialogue, you can opt to receive these URLs by setting this parameter to `true`."
14455
+ ),
14456
+ override_audio_redaction_method: import_zod3.z.enum(["silence"]).optional().describe(
14457
+ "Specify the method used to redact audio. By default, redacted audio uses a beep sound. Set to `silence` to replace PII with silence instead of a beep."
14458
+ )
14459
+ }).optional().describe(
14460
+ "The options for PII-redacted audio, if redact_pii_audio is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more information.\n"
13560
14461
  ),
13561
14462
  redact_pii_audio_quality: import_zod3.z.enum(["mp3", "wav"]).describe(
13562
- "Controls the filetype of the audio created by redact_pii_audio. Currently supports mp3 (default) and wav. See [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more details."
14463
+ "Controls the filetype of the audio created by redact_pii_audio. Currently supports mp3 (default) and wav. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more details."
13563
14464
  ).or(import_zod3.z.null()).optional().describe(
13564
- "The audio quality of the PII-redacted audio file, if redact_pii_audio is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more information.\n"
14465
+ "The audio quality of the PII-redacted audio file, if redact_pii_audio is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more information.\n"
13565
14466
  ),
13566
14467
  redact_pii_policies: import_zod3.z.array(
13567
14468
  import_zod3.z.enum([
@@ -13611,271 +14512,256 @@ var deleteTranscriptResponse = import_zod3.z.object({
13611
14512
  "zodiac_sign"
13612
14513
  ]).describe("The type of PII to redact")
13613
14514
  ).nullish().describe(
13614
- "The list of PII Redaction policies that were enabled, if PII Redaction is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more information.\n"
14515
+ "The list of PII Redaction policies that were enabled, if PII Redaction is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/pii-redaction) for more information.\n"
13615
14516
  ),
13616
14517
  redact_pii_sub: import_zod3.z.enum(["entity_name", "hash"]).optional().describe(
13617
- 'The replacement logic for detected PII, can be "entity_name" or "hash". See [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more details.'
14518
+ "The replacement logic for detected PII, can be `entity_name` or `hash`. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction) for more details."
14519
+ ),
14520
+ sentiment_analysis: import_zod3.z.boolean().nullish().describe(
14521
+ "Whether [Sentiment Analysis](https://www.assemblyai.com/docs/speech-understanding/sentiment-analysis) is enabled, can be true or false"
14522
+ ),
14523
+ sentiment_analysis_results: import_zod3.z.array(
14524
+ import_zod3.z.object({
14525
+ text: import_zod3.z.string().describe("The transcript of the sentence"),
14526
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, of the sentence"),
14527
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, of the sentence"),
14528
+ sentiment: import_zod3.z.enum(["POSITIVE", "NEUTRAL", "NEGATIVE"]),
14529
+ confidence: import_zod3.z.number().describe(
14530
+ "The confidence score for the detected sentiment of the sentence, from 0 to 1"
14531
+ ),
14532
+ channel: import_zod3.z.string().nullish().describe(
14533
+ "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
14534
+ ),
14535
+ speaker: import_zod3.z.string().nullable().describe(
14536
+ "The speaker of the sentence if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
14537
+ )
14538
+ }).describe("The result of the Sentiment Analysis model")
14539
+ ).nullish().describe(
14540
+ "An array of results for the Sentiment Analysis model, if it is enabled.\nSee [Sentiment Analysis](https://www.assemblyai.com/docs/speech-understanding/sentiment-analysis) for more information.\n"
13618
14541
  ),
13619
14542
  speaker_labels: import_zod3.z.boolean().nullish().describe(
13620
- "Whether [Speaker diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, can be true or false"
14543
+ "Whether [Speaker diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, can be true or false"
13621
14544
  ),
13622
14545
  speakers_expected: import_zod3.z.number().nullish().describe(
13623
- "Tell the speaker label model how many speakers it should attempt to identify. See [Speaker diarization](https://www.assemblyai.com/docs/models/speaker-diarization) for more details."
14546
+ "Tell the speaker label model how many speakers it should attempt to identify. See [Set number of speakers expected](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization#set-number-of-speakers-expected) for more details."
13624
14547
  ),
13625
- content_safety: import_zod3.z.boolean().nullish().describe(
13626
- "Whether [Content Moderation](https://www.assemblyai.com/docs/models/content-moderation) is enabled, can be true or false"
14548
+ speech_model_used: import_zod3.z.string().optional().describe(
14549
+ "The speech model to use for the transcription. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models."
13627
14550
  ),
13628
- content_safety_labels: import_zod3.z.object({
13629
- status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
13630
- results: import_zod3.z.array(
13631
- import_zod3.z.object({
13632
- text: import_zod3.z.string().describe("The transcript of the section flagged by the Content Moderation model"),
13633
- labels: import_zod3.z.array(
13634
- import_zod3.z.object({
13635
- label: import_zod3.z.string().describe("The label of the sensitive topic"),
13636
- confidence: import_zod3.z.number().min(
13637
- deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin
13638
- ).max(
13639
- deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax
13640
- ).describe("The confidence score for the topic being discussed, from 0 to 1"),
13641
- severity: import_zod3.z.number().min(
13642
- deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin
13643
- ).max(
13644
- deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax
13645
- ).describe("How severely the topic is discussed in the section, from 0 to 1")
13646
- })
13647
- ).describe(
13648
- "An array of safety labels, one per sensitive topic that was detected in the section"
14551
+ speech_models: import_zod3.z.array(
14552
+ import_zod3.z.string().describe(
14553
+ "The speech model to use for the transcription. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models."
14554
+ )
14555
+ ).nullish().describe(
14556
+ "List multiple speech models in priority order, allowing our system to automatically route your audio to the best available option. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models and routing behavior.\n"
14557
+ ),
14558
+ speech_threshold: import_zod3.z.number().nullish().describe(
14559
+ "Defaults to null. Reject audio files that contain less than this fraction of speech.\nValid values are in the range [0, 1] inclusive. See [Speech Threshold](https://www.assemblyai.com/docs/speech-threshold) for more details.\n"
14560
+ ),
14561
+ speech_understanding: import_zod3.z.object({
14562
+ request: import_zod3.z.object({
14563
+ translation: import_zod3.z.object({
14564
+ target_languages: import_zod3.z.array(import_zod3.z.string()).describe(
14565
+ 'List of target language codes (e.g., `["es", "de"]`). See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for supported languages.'
13649
14566
  ),
13650
- sentences_idx_start: import_zod3.z.number().describe("The sentence index at which the section begins"),
13651
- sentences_idx_end: import_zod3.z.number().describe("The sentence index at which the section ends"),
13652
- timestamp: import_zod3.z.object({
13653
- start: import_zod3.z.number().describe("The start time in milliseconds"),
13654
- end: import_zod3.z.number().describe("The end time in milliseconds")
13655
- }).describe("Timestamp containing a start and end property in milliseconds")
14567
+ formal: import_zod3.z.boolean().default(deleteTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault).describe(
14568
+ "Use formal language style. See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for more details."
14569
+ ),
14570
+ match_original_utterance: import_zod3.z.boolean().optional().describe(
14571
+ "When enabled with Speaker Labels, returns translated text in the utterances array. Each utterance will include a `translated_texts` key containing translations for each target language."
14572
+ )
13656
14573
  })
13657
- ).describe("An array of results for the Content Moderation model"),
13658
- summary: import_zod3.z.record(
13659
- import_zod3.z.string(),
13660
- import_zod3.z.number().min(deleteTranscriptResponseContentSafetyLabelsSummaryMinOne).max(deleteTranscriptResponseContentSafetyLabelsSummaryMaxOne).describe(
13661
- 'A confidence score for the presence of the sensitive topic "topic" across the entire audio file'
14574
+ }).describe(
14575
+ "Request body for [Translation](https://www.assemblyai.com/docs/speech-understanding/translation)."
14576
+ ).or(
14577
+ import_zod3.z.object({
14578
+ speaker_identification: import_zod3.z.object({
14579
+ speaker_type: import_zod3.z.enum(["role", "name"]).describe(
14580
+ "Type of speaker identification. See [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification) for details on each type."
14581
+ ),
14582
+ known_values: import_zod3.z.array(import_zod3.z.string()).optional().describe(
14583
+ 'Required if speaker_type is "role". Each value must be 35 characters or less.'
14584
+ ),
14585
+ speakers: import_zod3.z.array(
14586
+ import_zod3.z.object({
14587
+ role: import_zod3.z.string().optional().describe(
14588
+ 'The role of the speaker. Required when `speaker_type` is "role".'
14589
+ ),
14590
+ name: import_zod3.z.string().optional().describe(
14591
+ 'The name of the speaker. Required when `speaker_type` is "name".'
14592
+ ),
14593
+ description: import_zod3.z.string().optional().describe(
14594
+ "A description of the speaker to help the model identify them based on conversational context."
14595
+ )
14596
+ })
14597
+ ).optional().describe(
14598
+ "An array of speaker objects with metadata to improve identification accuracy. Each object should include a `role` or `name` (depending on `speaker_type`) and an optional `description` to help the model identify the speaker. You can also include any additional custom properties (e.g., `company`, `title`) to provide more context. Use this as an alternative to `known_values` when you want to provide additional context about each speaker."
14599
+ )
14600
+ })
14601
+ }).describe(
14602
+ "Request body for [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification)."
13662
14603
  )
13663
- ).describe(
13664
- "A summary of the Content Moderation confidence results for the entire audio file"
13665
- ),
13666
- severity_score_summary: import_zod3.z.record(
13667
- import_zod3.z.string(),
14604
+ ).or(
14605
+ import_zod3.z.object({
14606
+ custom_formatting: import_zod3.z.object({
14607
+ date: import_zod3.z.string().optional().describe(
14608
+ 'Date format pattern (e.g., `"mm/dd/yyyy"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
14609
+ ),
14610
+ phone_number: import_zod3.z.string().optional().describe(
14611
+ 'Phone number format pattern (e.g., `"(xxx)xxx-xxxx"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
14612
+ ),
14613
+ email: import_zod3.z.string().optional().describe(
14614
+ 'Email format pattern (e.g., `"username@domain.com"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
14615
+ )
14616
+ })
14617
+ }).describe(
14618
+ "Request body for [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting)."
14619
+ )
14620
+ ).optional(),
14621
+ response: import_zod3.z.object({
14622
+ translation: import_zod3.z.object({
14623
+ status: import_zod3.z.string().optional()
14624
+ }).optional()
14625
+ }).or(
13668
14626
  import_zod3.z.object({
13669
- low: import_zod3.z.number().min(deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin).max(deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax),
13670
- medium: import_zod3.z.number().min(deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin).max(deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax),
13671
- high: import_zod3.z.number().min(deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin).max(deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax)
14627
+ speaker_identification: import_zod3.z.object({
14628
+ mapping: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional().describe(
14629
+ 'A mapping of the original generic speaker labels (e.g., "A", "B") to the identified speaker names or roles.'
14630
+ ),
14631
+ status: import_zod3.z.string().optional()
14632
+ }).optional()
13672
14633
  })
13673
- ).describe(
13674
- "A summary of the Content Moderation severity results for the entire audio file"
13675
- )
13676
- }).describe(
13677
- "An array of results for the Content Moderation model, if it is enabled.\nSee [Content moderation](https://www.assemblyai.com/docs/models/content-moderation) for more information.\n"
13678
- ).or(import_zod3.z.null()).optional().describe(
13679
- "An array of results for the Content Moderation model, if it is enabled.\nSee [Content moderation](https://www.assemblyai.com/docs/models/content-moderation) for more information.\n"
13680
- ),
13681
- iab_categories: import_zod3.z.boolean().nullish().describe(
13682
- "Whether [Topic Detection](https://www.assemblyai.com/docs/models/topic-detection) is enabled, can be true or false"
13683
- ),
13684
- iab_categories_result: import_zod3.z.object({
13685
- status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
13686
- results: import_zod3.z.array(
14634
+ ).or(
13687
14635
  import_zod3.z.object({
13688
- text: import_zod3.z.string().describe("The text in the transcript in which a detected topic occurs"),
13689
- labels: import_zod3.z.array(
13690
- import_zod3.z.object({
13691
- relevance: import_zod3.z.number().min(
13692
- deleteTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin
13693
- ).max(
13694
- deleteTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax
13695
- ).describe("How relevant the detected topic is of a detected topic"),
13696
- label: import_zod3.z.string().describe(
13697
- "The IAB taxonomical label for the label of the detected topic, where > denotes supertopic/subtopic relationship"
13698
- )
13699
- })
13700
- ).optional().describe("An array of detected topics in the text"),
13701
- timestamp: import_zod3.z.object({
13702
- start: import_zod3.z.number().describe("The start time in milliseconds"),
13703
- end: import_zod3.z.number().describe("The end time in milliseconds")
13704
- }).optional().describe("Timestamp containing a start and end property in milliseconds")
13705
- }).describe("The result of the topic detection model")
13706
- ).describe("An array of results for the Topic Detection model"),
13707
- summary: import_zod3.z.record(
13708
- import_zod3.z.string(),
13709
- import_zod3.z.number().min(deleteTranscriptResponseIabCategoriesResultSummaryMinOne).max(deleteTranscriptResponseIabCategoriesResultSummaryMaxOne)
13710
- ).describe("The overall relevance of topic to the entire audio file")
13711
- }).describe(
13712
- "The result of the Topic Detection model, if it is enabled.\nSee [Topic Detection](https://www.assemblyai.com/docs/models/topic-detection) for more information.\n"
13713
- ).or(import_zod3.z.null()).optional().describe(
13714
- "The result of the Topic Detection model, if it is enabled.\nSee [Topic Detection](https://www.assemblyai.com/docs/models/topic-detection) for more information.\n"
13715
- ),
13716
- custom_spelling: import_zod3.z.array(
13717
- import_zod3.z.object({
13718
- from: import_zod3.z.array(import_zod3.z.string().describe("Word or phrase to replace")).describe("Words or phrases to replace"),
13719
- to: import_zod3.z.string().describe("Word to replace with")
13720
- }).describe(
13721
- "Object containing words or phrases to replace, and the word or phrase to replace with"
13722
- )
13723
- ).nullish().describe("Customize how words are spelled and formatted using to and from values"),
13724
- keyterms_prompt: import_zod3.z.array(import_zod3.z.string()).optional().describe(
13725
- "Improve accuracy with up to 1000 domain-specific words or phrases (maximum 6 words per phrase).\n"
14636
+ custom_formatting: import_zod3.z.object({
14637
+ mapping: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional(),
14638
+ formatted_text: import_zod3.z.string().optional()
14639
+ }).optional()
14640
+ })
14641
+ ).optional()
14642
+ }).optional().describe(
14643
+ "Speech understanding tasks like [Translation](https://www.assemblyai.com/docs/speech-understanding/translation), [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification), and [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting). See the task-specific docs for available options and configuration.\n"
13726
14644
  ),
13727
- prompt: import_zod3.z.string().optional().describe("This parameter does not currently have any functionality attached to it."),
13728
- auto_chapters: import_zod3.z.boolean().nullish().describe(
13729
- "Whether [Auto Chapters](https://www.assemblyai.com/docs/models/auto-chapters) is enabled, can be true or false"
14645
+ status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).describe(
14646
+ "The status of your transcript. Possible values are queued, processing, completed, or error."
13730
14647
  ),
13731
- chapters: import_zod3.z.array(
13732
- import_zod3.z.object({
13733
- gist: import_zod3.z.string().describe(
13734
- "An ultra-short summary (just a few words) of the content spoken in the chapter"
13735
- ),
13736
- headline: import_zod3.z.string().describe("A single sentence summary of the content spoken during the chapter"),
13737
- summary: import_zod3.z.string().describe("A one paragraph summary of the content spoken during the chapter"),
13738
- start: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter"),
13739
- end: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter")
13740
- }).describe("Chapter of the audio file")
13741
- ).nullish().describe("An array of temporally sequential chapters for the audio file"),
13742
14648
  summarization: import_zod3.z.boolean().describe(
13743
- "Whether [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled, either true or false"
14649
+ "Whether [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization) is enabled, either true or false. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details.\n\nNote: This parameter is only supported for the Universal-2 model.\n"
13744
14650
  ),
13745
- summary_type: import_zod3.z.string().nullish().describe(
13746
- "The type of summary generated, if [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled"
14651
+ summary: import_zod3.z.string().nullish().describe(
14652
+ "The generated summary of the media file, if [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization) is enabled. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details."
13747
14653
  ),
13748
14654
  summary_model: import_zod3.z.string().nullish().describe(
13749
- "The Summarization model used to generate the summary,\nif [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled\n"
14655
+ "The Summarization model used to generate the summary,\nif [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization#summary-models) is enabled. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details.\n"
13750
14656
  ),
13751
- summary: import_zod3.z.string().nullish().describe(
13752
- "The generated summary of the media file, if [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled"
14657
+ summary_type: import_zod3.z.string().nullish().describe(
14658
+ "The type of summary generated, if [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization#summary-types) is enabled. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details."
13753
14659
  ),
13754
- custom_topics: import_zod3.z.boolean().nullish().describe("Whether custom topics is enabled, either true or false"),
13755
- topics: import_zod3.z.array(import_zod3.z.string()).optional().describe("The list of custom topics provided if custom topics is enabled"),
13756
- sentiment_analysis: import_zod3.z.boolean().nullish().describe(
13757
- "Whether [Sentiment Analysis](https://www.assemblyai.com/docs/models/sentiment-analysis) is enabled, can be true or false"
14660
+ remove_audio_tags: import_zod3.z.enum(["all"]).describe(
14661
+ "Whether [audio event tags](https://www.assemblyai.com/docs/pre-recorded-audio/universal-3-pro#audio-event-tags) were removed from the transcript text.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n"
14662
+ ).or(import_zod3.z.null()).optional().describe(
14663
+ "Whether [audio event tags](https://www.assemblyai.com/docs/pre-recorded-audio/universal-3-pro#audio-event-tags) were removed from the transcript text.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n"
13758
14664
  ),
13759
- sentiment_analysis_results: import_zod3.z.array(
14665
+ temperature: import_zod3.z.number().nullish().describe(
14666
+ "The temperature that was used for the model's response. See the [Prompting Guide](https://www.assemblyai.com/docs/pre-recorded-audio/prompting) for more details.\n\nNote: This parameter can only be used with the Universal-3 Pro model.\n"
14667
+ ),
14668
+ text: import_zod3.z.string().nullish().describe("The textual transcript of your media file"),
14669
+ throttled: import_zod3.z.boolean().nullish().describe(
14670
+ "True while a request is throttled and false when a request is no longer throttled"
14671
+ ),
14672
+ utterances: import_zod3.z.array(
13760
14673
  import_zod3.z.object({
13761
- text: import_zod3.z.string().describe("The transcript of the sentence"),
13762
- start: import_zod3.z.number().describe("The starting time, in milliseconds, of the sentence"),
13763
- end: import_zod3.z.number().describe("The ending time, in milliseconds, of the sentence"),
13764
- sentiment: import_zod3.z.enum(["POSITIVE", "NEUTRAL", "NEGATIVE"]),
13765
- confidence: import_zod3.z.number().min(deleteTranscriptResponseSentimentAnalysisResultsItemConfidenceMin).max(deleteTranscriptResponseSentimentAnalysisResultsItemConfidenceMax).describe(
13766
- "The confidence score for the detected sentiment of the sentence, from 0 to 1"
13767
- ),
14674
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this utterance"),
14675
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, of the utterance in the audio file"),
14676
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, of the utterance in the audio file"),
14677
+ text: import_zod3.z.string().describe("The text for this utterance"),
14678
+ words: import_zod3.z.array(
14679
+ import_zod3.z.object({
14680
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
14681
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
14682
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
14683
+ text: import_zod3.z.string().describe("The text of the word"),
14684
+ channel: import_zod3.z.string().nullish().describe(
14685
+ "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
14686
+ ),
14687
+ speaker: import_zod3.z.string().nullable().describe(
14688
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
14689
+ )
14690
+ })
14691
+ ).describe("The words in the utterance."),
13768
14692
  channel: import_zod3.z.string().nullish().describe(
13769
14693
  "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13770
14694
  ),
13771
- speaker: import_zod3.z.string().nullable().describe(
13772
- "The speaker of the sentence if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
14695
+ speaker: import_zod3.z.string().describe(
14696
+ 'The speaker of this utterance, where each speaker is assigned a sequential capital letter - e.g. "A" for Speaker A, "B" for Speaker B, etc.'
14697
+ ),
14698
+ translated_texts: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional().describe(
14699
+ 'Translations keyed by language code (e.g., `{"es": "Texto traducido", "de": "\xDCbersetzter Text"}`). Only present when `match_original_utterance` is enabled with translation.'
13773
14700
  )
13774
- }).describe("The result of the Sentiment Analysis model")
14701
+ })
13775
14702
  ).nullish().describe(
13776
- "An array of results for the Sentiment Analysis model, if it is enabled.\nSee [Sentiment Analysis](https://www.assemblyai.com/docs/models/sentiment-analysis) for more information.\n"
14703
+ "When multichannel or speaker_labels is enabled, a list of turn-by-turn utterance objects.\nSee [Speaker diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) and [Multichannel transcription](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) for more information.\n"
13777
14704
  ),
13778
- entity_detection: import_zod3.z.boolean().nullish().describe(
13779
- "Whether [Entity Detection](https://www.assemblyai.com/docs/models/entity-detection) is enabled, can be true or false"
14705
+ webhook_auth: import_zod3.z.boolean().describe(
14706
+ "Whether [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) authentication details were provided"
13780
14707
  ),
13781
- entities: import_zod3.z.array(
14708
+ webhook_auth_header_name: import_zod3.z.string().nullish().describe(
14709
+ "The header name to be sent with the transcript completed or failed [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) requests"
14710
+ ),
14711
+ webhook_status_code: import_zod3.z.number().nullish().describe(
14712
+ "The status code we received from your server when delivering the transcript completed or failed [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) request, if a webhook URL was provided"
14713
+ ),
14714
+ webhook_url: import_zod3.z.string().nullish().describe(
14715
+ "The URL to which we send [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) requests.\n"
14716
+ ),
14717
+ words: import_zod3.z.array(
13782
14718
  import_zod3.z.object({
13783
- entity_type: import_zod3.z.enum([
13784
- "account_number",
13785
- "banking_information",
13786
- "blood_type",
13787
- "credit_card_cvv",
13788
- "credit_card_expiration",
13789
- "credit_card_number",
13790
- "date",
13791
- "date_interval",
13792
- "date_of_birth",
13793
- "drivers_license",
13794
- "drug",
13795
- "duration",
13796
- "email_address",
13797
- "event",
13798
- "filename",
13799
- "gender_sexuality",
13800
- "healthcare_number",
13801
- "injury",
13802
- "ip_address",
13803
- "language",
13804
- "location",
13805
- "marital_status",
13806
- "medical_condition",
13807
- "medical_process",
13808
- "money_amount",
13809
- "nationality",
13810
- "number_sequence",
13811
- "occupation",
13812
- "organization",
13813
- "passport_number",
13814
- "password",
13815
- "person_age",
13816
- "person_name",
13817
- "phone_number",
13818
- "physical_attribute",
13819
- "political_affiliation",
13820
- "religion",
13821
- "statistics",
13822
- "time",
13823
- "url",
13824
- "us_social_security_number",
13825
- "username",
13826
- "vehicle_id",
13827
- "zodiac_sign"
13828
- ]).describe("The type of entity for the detected entity"),
13829
- text: import_zod3.z.string().describe("The text for the detected entity"),
13830
- start: import_zod3.z.number().describe(
13831
- "The starting time, in milliseconds, at which the detected entity appears in the audio file"
14719
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
14720
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
14721
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
14722
+ text: import_zod3.z.string().describe("The text of the word"),
14723
+ channel: import_zod3.z.string().nullish().describe(
14724
+ "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13832
14725
  ),
13833
- end: import_zod3.z.number().describe(
13834
- "The ending time, in milliseconds, for the detected entity in the audio file"
14726
+ speaker: import_zod3.z.string().nullable().describe(
14727
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
13835
14728
  )
13836
- }).describe("A detected entity")
14729
+ })
13837
14730
  ).nullish().describe(
13838
- "An array of results for the Entity Detection model, if it is enabled.\nSee [Entity detection](https://www.assemblyai.com/docs/models/entity-detection) for more information.\n"
13839
- ),
13840
- speech_threshold: import_zod3.z.number().min(deleteTranscriptResponseSpeechThresholdMin).max(deleteTranscriptResponseSpeechThresholdMax).nullish().describe(
13841
- "Defaults to null. Reject audio files that contain less than this fraction of speech.\nValid values are in the range [0, 1] inclusive.\n"
13842
- ),
13843
- throttled: import_zod3.z.boolean().nullish().describe(
13844
- "True while a request is throttled and false when a request is no longer throttled"
13845
- ),
13846
- error: import_zod3.z.string().optional().describe("Error message of why the transcript failed"),
13847
- language_model: import_zod3.z.string().describe("The language model that was used for the transcript"),
13848
- acoustic_model: import_zod3.z.string().describe("The acoustic model that was used for the transcript")
14731
+ "An array of temporally-sequential word objects, one for each word in the transcript.\n"
14732
+ ),
14733
+ acoustic_model: import_zod3.z.string().describe("This parameter does not currently have any functionality attached to it."),
14734
+ custom_topics: import_zod3.z.boolean().nullish().describe("This parameter does not currently have any functionality attached to it."),
14735
+ language_model: import_zod3.z.string().describe("This parameter does not currently have any functionality attached to it."),
14736
+ speech_model: import_zod3.z.string().describe(
14737
+ "The speech model to use for the transcription. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models."
14738
+ ).or(import_zod3.z.null()).describe(
14739
+ "This parameter has been replaced with the `speech_models` parameter, learn more about the `speech_models` parameter [here](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model).\n"
14740
+ ),
14741
+ speed_boost: import_zod3.z.boolean().nullish().describe("This parameter does not currently have any functionality attached to it."),
14742
+ topics: import_zod3.z.array(import_zod3.z.string()).optional().describe("This parameter does not currently have any functionality attached to it."),
14743
+ translated_texts: import_zod3.z.object({
14744
+ language_code: import_zod3.z.string().optional().describe("Translated text for this language code")
14745
+ }).optional().describe(
14746
+ "Translated text keyed by language code. See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for more details."
14747
+ )
13849
14748
  }).describe("A transcript object");
13850
- var getSubtitlesParams = import_zod3.z.object({
13851
- transcript_id: import_zod3.z.string().describe("ID of the transcript"),
13852
- subtitle_format: import_zod3.z.enum(["srt", "vtt"]).describe("The format of the captions")
13853
- });
13854
- var getSubtitlesQueryParams = import_zod3.z.object({
13855
- chars_per_caption: import_zod3.z.number().optional().describe("The maximum number of characters per caption")
13856
- });
13857
14749
  var getTranscriptSentencesParams = import_zod3.z.object({
13858
14750
  transcript_id: import_zod3.z.string().describe("ID of the transcript")
13859
14751
  });
13860
- var getTranscriptSentencesResponseConfidenceMin = 0;
13861
- var getTranscriptSentencesResponseConfidenceMax = 1;
13862
- var getTranscriptSentencesResponseSentencesItemConfidenceMin = 0;
13863
- var getTranscriptSentencesResponseSentencesItemConfidenceMax = 1;
13864
- var getTranscriptSentencesResponseSentencesItemWordsItemConfidenceMin = 0;
13865
- var getTranscriptSentencesResponseSentencesItemWordsItemConfidenceMax = 1;
13866
14752
  var getTranscriptSentencesResponse = import_zod3.z.object({
13867
14753
  id: import_zod3.z.string().uuid().describe("The unique identifier for the transcript"),
13868
- confidence: import_zod3.z.number().min(getTranscriptSentencesResponseConfidenceMin).max(getTranscriptSentencesResponseConfidenceMax).describe("The confidence score for the transcript"),
14754
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript"),
13869
14755
  audio_duration: import_zod3.z.number().describe("The duration of the audio file in seconds"),
13870
14756
  sentences: import_zod3.z.array(
13871
14757
  import_zod3.z.object({
13872
14758
  text: import_zod3.z.string().describe("The transcript of the sentence"),
13873
14759
  start: import_zod3.z.number().describe("The starting time, in milliseconds, for the sentence"),
13874
14760
  end: import_zod3.z.number().describe("The ending time, in milliseconds, for the sentence"),
13875
- confidence: import_zod3.z.number().min(getTranscriptSentencesResponseSentencesItemConfidenceMin).max(getTranscriptSentencesResponseSentencesItemConfidenceMax).describe("The confidence score for the transcript of this sentence"),
14761
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this sentence"),
13876
14762
  words: import_zod3.z.array(
13877
14763
  import_zod3.z.object({
13878
- confidence: import_zod3.z.number().min(getTranscriptSentencesResponseSentencesItemWordsItemConfidenceMin).max(getTranscriptSentencesResponseSentencesItemWordsItemConfidenceMax).describe("The confidence score for the transcript of this word"),
14764
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
13879
14765
  start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
13880
14766
  end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
13881
14767
  text: import_zod3.z.string().describe("The text of the word"),
@@ -13883,7 +14769,7 @@ var getTranscriptSentencesResponse = import_zod3.z.object({
13883
14769
  "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13884
14770
  ),
13885
14771
  speaker: import_zod3.z.string().nullable().describe(
13886
- "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
14772
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
13887
14773
  )
13888
14774
  })
13889
14775
  ).describe("An array of words in the sentence"),
@@ -13891,7 +14777,7 @@ var getTranscriptSentencesResponse = import_zod3.z.object({
13891
14777
  "The channel of the sentence. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13892
14778
  ),
13893
14779
  speaker: import_zod3.z.string().nullable().describe(
13894
- "The speaker of the sentence if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
14780
+ "The speaker of the sentence if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
13895
14781
  )
13896
14782
  })
13897
14783
  ).describe("An array of sentences in the transcript")
@@ -13899,25 +14785,19 @@ var getTranscriptSentencesResponse = import_zod3.z.object({
13899
14785
  var getTranscriptParagraphsParams = import_zod3.z.object({
13900
14786
  transcript_id: import_zod3.z.string().describe("ID of the transcript")
13901
14787
  });
13902
- var getTranscriptParagraphsResponseConfidenceMin = 0;
13903
- var getTranscriptParagraphsResponseConfidenceMax = 1;
13904
- var getTranscriptParagraphsResponseParagraphsItemConfidenceMin = 0;
13905
- var getTranscriptParagraphsResponseParagraphsItemConfidenceMax = 1;
13906
- var getTranscriptParagraphsResponseParagraphsItemWordsItemConfidenceMin = 0;
13907
- var getTranscriptParagraphsResponseParagraphsItemWordsItemConfidenceMax = 1;
13908
14788
  var getTranscriptParagraphsResponse = import_zod3.z.object({
13909
14789
  id: import_zod3.z.string().uuid().describe("The unique identifier of your transcript"),
13910
- confidence: import_zod3.z.number().min(getTranscriptParagraphsResponseConfidenceMin).max(getTranscriptParagraphsResponseConfidenceMax).describe("The confidence score for the transcript"),
14790
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript"),
13911
14791
  audio_duration: import_zod3.z.number().describe("The duration of the audio file in seconds"),
13912
14792
  paragraphs: import_zod3.z.array(
13913
14793
  import_zod3.z.object({
13914
14794
  text: import_zod3.z.string().describe("The transcript of the paragraph"),
13915
14795
  start: import_zod3.z.number().describe("The starting time, in milliseconds, of the paragraph"),
13916
14796
  end: import_zod3.z.number().describe("The ending time, in milliseconds, of the paragraph"),
13917
- confidence: import_zod3.z.number().min(getTranscriptParagraphsResponseParagraphsItemConfidenceMin).max(getTranscriptParagraphsResponseParagraphsItemConfidenceMax).describe("The confidence score for the transcript of this paragraph"),
14797
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this paragraph"),
13918
14798
  words: import_zod3.z.array(
13919
14799
  import_zod3.z.object({
13920
- confidence: import_zod3.z.number().min(getTranscriptParagraphsResponseParagraphsItemWordsItemConfidenceMin).max(getTranscriptParagraphsResponseParagraphsItemWordsItemConfidenceMax).describe("The confidence score for the transcript of this word"),
14800
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
13921
14801
  start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
13922
14802
  end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
13923
14803
  text: import_zod3.z.string().describe("The text of the word"),
@@ -13925,13 +14805,28 @@ var getTranscriptParagraphsResponse = import_zod3.z.object({
13925
14805
  "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13926
14806
  ),
13927
14807
  speaker: import_zod3.z.string().nullable().describe(
13928
- "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
14808
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
13929
14809
  )
13930
14810
  })
13931
14811
  ).describe("An array of words in the paragraph")
13932
14812
  })
13933
14813
  ).describe("An array of paragraphs in the transcript")
13934
14814
  });
14815
+ var getSubtitlesParams = import_zod3.z.object({
14816
+ transcript_id: import_zod3.z.string().describe("ID of the transcript"),
14817
+ subtitle_format: import_zod3.z.enum(["srt", "vtt"]).describe("The format of the captions")
14818
+ });
14819
+ var getSubtitlesQueryParams = import_zod3.z.object({
14820
+ chars_per_caption: import_zod3.z.number().optional().describe("The maximum number of characters per caption")
14821
+ });
14822
+ var getSubtitlesResponse = import_zod3.z.object({});
14823
+ var getRedactedAudioParams = import_zod3.z.object({
14824
+ transcript_id: import_zod3.z.string().describe("ID of the transcript")
14825
+ });
14826
+ var getRedactedAudioResponse = import_zod3.z.object({
14827
+ status: import_zod3.z.enum(["redacted_audio_ready"]).describe("The status of the redacted audio"),
14828
+ redacted_audio_url: import_zod3.z.string().describe("The URL of the redacted audio file")
14829
+ });
13935
14830
  var wordSearchParams = import_zod3.z.object({
13936
14831
  transcript_id: import_zod3.z.string().describe("ID of the transcript")
13937
14832
  });
@@ -13948,7 +14843,7 @@ var wordSearchResponse = import_zod3.z.object({
13948
14843
  text: import_zod3.z.string().describe("The matched word"),
13949
14844
  count: import_zod3.z.number().describe("The total amount of times the word is in the transcript"),
13950
14845
  timestamps: import_zod3.z.array(
13951
- import_zod3.z.array(import_zod3.z.number().describe("Timestamp in milliseconds")).describe(
14846
+ import_zod3.z.array(import_zod3.z.number()).describe(
13952
14847
  "An array of timestamps structured as [`start_time`, `end_time`] in milliseconds"
13953
14848
  )
13954
14849
  ).describe("An array of timestamps"),
@@ -13958,205 +14853,6 @@ var wordSearchResponse = import_zod3.z.object({
13958
14853
  })
13959
14854
  ).describe("The matches of the search")
13960
14855
  });
13961
- var getRedactedAudioParams = import_zod3.z.object({
13962
- transcript_id: import_zod3.z.string().describe("ID of the transcript")
13963
- });
13964
- var getRedactedAudioResponse = import_zod3.z.object({
13965
- status: import_zod3.z.enum(["redacted_audio_ready"]).describe("The status of the redacted audio"),
13966
- redacted_audio_url: import_zod3.z.string().describe("The URL of the redacted audio file")
13967
- });
13968
- var createTemporaryTokenBodyExpiresInMin = 60;
13969
- var createTemporaryTokenBody = import_zod3.z.object({
13970
- expires_in: import_zod3.z.number().min(createTemporaryTokenBodyExpiresInMin).describe("The amount of time until the token expires in seconds")
13971
- });
13972
- var createTemporaryTokenResponse = import_zod3.z.object({
13973
- token: import_zod3.z.string().describe("The temporary authentication token for Streaming Speech-to-Text")
13974
- });
13975
- var lemurTaskBodyFinalModelDefault = "default";
13976
- var lemurTaskBodyMaxOutputSizeDefault = 2e3;
13977
- var lemurTaskBodyTemperatureDefault = 0;
13978
- var lemurTaskBodyTemperatureMin = 0;
13979
- var lemurTaskBodyTemperatureMax = 1;
13980
- var lemurTaskBody = import_zod3.z.object({
13981
- prompt: import_zod3.z.string().describe(
13982
- "Your text to prompt the model to produce a desired output, including any context you want to pass into the model."
13983
- )
13984
- }).and(
13985
- import_zod3.z.object({
13986
- transcript_ids: import_zod3.z.array(import_zod3.z.string().uuid()).optional().describe(
13987
- "A list of completed transcripts with text. Up to a maximum of 100 hours of audio.\nUse either transcript_ids or input_text as input into LeMUR.\n"
13988
- ),
13989
- input_text: import_zod3.z.string().optional().describe(
13990
- "Custom formatted transcript data. Maximum size is the context limit of the selected model.\nUse either transcript_ids or input_text as input into LeMUR.\n"
13991
- ),
13992
- context: import_zod3.z.string().or(import_zod3.z.record(import_zod3.z.string(), import_zod3.z.any())).optional().describe("Context to provide the model. This can be a string or a free-form JSON value."),
13993
- final_model: import_zod3.z.enum([
13994
- "anthropic/claude-3-5-sonnet",
13995
- "anthropic/claude-3-opus",
13996
- "anthropic/claude-3-haiku"
13997
- ]).describe("The model that is used for the final prompt after compression is performed.\n").or(import_zod3.z.string()).describe("The model that is used for the final prompt after compression is performed.\n"),
13998
- max_output_size: import_zod3.z.number().default(lemurTaskBodyMaxOutputSizeDefault).describe("Max output size in tokens."),
13999
- temperature: import_zod3.z.number().min(lemurTaskBodyTemperatureMin).max(lemurTaskBodyTemperatureMax).optional().describe(
14000
- "The temperature to use for the model.\nHigher values result in answers that are more creative, lower values are more conservative.\nCan be any value between 0.0 and 1.0 inclusive.\n"
14001
- )
14002
- })
14003
- );
14004
- var lemurTaskResponseUsageInputTokensMin = 0;
14005
- var lemurTaskResponseUsageOutputTokensMin = 0;
14006
- var lemurTaskResponse = import_zod3.z.object({
14007
- response: import_zod3.z.string().describe("The response generated by LeMUR.")
14008
- }).and(
14009
- import_zod3.z.object({
14010
- request_id: import_zod3.z.string().uuid().describe("The ID of the LeMUR request"),
14011
- usage: import_zod3.z.object({
14012
- input_tokens: import_zod3.z.number().min(lemurTaskResponseUsageInputTokensMin).describe("The number of input tokens used by the model"),
14013
- output_tokens: import_zod3.z.number().min(lemurTaskResponseUsageOutputTokensMin).describe("The number of output tokens generated by the model")
14014
- }).describe("The usage numbers for the LeMUR request")
14015
- })
14016
- );
14017
- var lemurSummaryBodyFinalModelDefault = "default";
14018
- var lemurSummaryBodyMaxOutputSizeDefault = 2e3;
14019
- var lemurSummaryBodyTemperatureDefault = 0;
14020
- var lemurSummaryBodyTemperatureMin = 0;
14021
- var lemurSummaryBodyTemperatureMax = 1;
14022
- var lemurSummaryBody = import_zod3.z.object({
14023
- transcript_ids: import_zod3.z.array(import_zod3.z.string().uuid()).optional().describe(
14024
- "A list of completed transcripts with text. Up to a maximum of 100 hours of audio.\nUse either transcript_ids or input_text as input into LeMUR.\n"
14025
- ),
14026
- input_text: import_zod3.z.string().optional().describe(
14027
- "Custom formatted transcript data. Maximum size is the context limit of the selected model.\nUse either transcript_ids or input_text as input into LeMUR.\n"
14028
- ),
14029
- context: import_zod3.z.string().or(import_zod3.z.record(import_zod3.z.string(), import_zod3.z.any())).optional().describe("Context to provide the model. This can be a string or a free-form JSON value."),
14030
- final_model: import_zod3.z.enum(["anthropic/claude-3-5-sonnet", "anthropic/claude-3-opus", "anthropic/claude-3-haiku"]).describe("The model that is used for the final prompt after compression is performed.\n").or(import_zod3.z.string()).describe("The model that is used for the final prompt after compression is performed.\n"),
14031
- max_output_size: import_zod3.z.number().default(lemurSummaryBodyMaxOutputSizeDefault).describe("Max output size in tokens."),
14032
- temperature: import_zod3.z.number().min(lemurSummaryBodyTemperatureMin).max(lemurSummaryBodyTemperatureMax).optional().describe(
14033
- "The temperature to use for the model.\nHigher values result in answers that are more creative, lower values are more conservative.\nCan be any value between 0.0 and 1.0 inclusive.\n"
14034
- )
14035
- }).and(
14036
- import_zod3.z.object({
14037
- answer_format: import_zod3.z.string().optional().describe(
14038
- 'How you want the summary to be returned. This can be any text. Examples: "TLDR", "bullet points"\n'
14039
- )
14040
- })
14041
- );
14042
- var lemurSummaryResponseUsageInputTokensMin = 0;
14043
- var lemurSummaryResponseUsageOutputTokensMin = 0;
14044
- var lemurSummaryResponse = import_zod3.z.object({
14045
- response: import_zod3.z.string().describe("The response generated by LeMUR.")
14046
- }).and(
14047
- import_zod3.z.object({
14048
- request_id: import_zod3.z.string().uuid().describe("The ID of the LeMUR request"),
14049
- usage: import_zod3.z.object({
14050
- input_tokens: import_zod3.z.number().min(lemurSummaryResponseUsageInputTokensMin).describe("The number of input tokens used by the model"),
14051
- output_tokens: import_zod3.z.number().min(lemurSummaryResponseUsageOutputTokensMin).describe("The number of output tokens generated by the model")
14052
- }).describe("The usage numbers for the LeMUR request")
14053
- })
14054
- );
14055
- var lemurQuestionAnswerBodyFinalModelDefault = "default";
14056
- var lemurQuestionAnswerBodyMaxOutputSizeDefault = 2e3;
14057
- var lemurQuestionAnswerBodyTemperatureDefault = 0;
14058
- var lemurQuestionAnswerBodyTemperatureMin = 0;
14059
- var lemurQuestionAnswerBodyTemperatureMax = 1;
14060
- var lemurQuestionAnswerBody = import_zod3.z.object({
14061
- transcript_ids: import_zod3.z.array(import_zod3.z.string().uuid()).optional().describe(
14062
- "A list of completed transcripts with text. Up to a maximum of 100 hours of audio.\nUse either transcript_ids or input_text as input into LeMUR.\n"
14063
- ),
14064
- input_text: import_zod3.z.string().optional().describe(
14065
- "Custom formatted transcript data. Maximum size is the context limit of the selected model.\nUse either transcript_ids or input_text as input into LeMUR.\n"
14066
- ),
14067
- context: import_zod3.z.string().or(import_zod3.z.record(import_zod3.z.string(), import_zod3.z.any())).optional().describe("Context to provide the model. This can be a string or a free-form JSON value."),
14068
- final_model: import_zod3.z.enum(["anthropic/claude-3-5-sonnet", "anthropic/claude-3-opus", "anthropic/claude-3-haiku"]).describe("The model that is used for the final prompt after compression is performed.\n").or(import_zod3.z.string()).describe("The model that is used for the final prompt after compression is performed.\n"),
14069
- max_output_size: import_zod3.z.number().default(lemurQuestionAnswerBodyMaxOutputSizeDefault).describe("Max output size in tokens."),
14070
- temperature: import_zod3.z.number().min(lemurQuestionAnswerBodyTemperatureMin).max(lemurQuestionAnswerBodyTemperatureMax).optional().describe(
14071
- "The temperature to use for the model.\nHigher values result in answers that are more creative, lower values are more conservative.\nCan be any value between 0.0 and 1.0 inclusive.\n"
14072
- )
14073
- }).and(
14074
- import_zod3.z.object({
14075
- questions: import_zod3.z.array(
14076
- import_zod3.z.object({
14077
- question: import_zod3.z.string().describe(
14078
- "The question you wish to ask. For more complex questions use default model."
14079
- ),
14080
- context: import_zod3.z.string().or(import_zod3.z.record(import_zod3.z.string(), import_zod3.z.any())).optional().describe(
14081
- "Any context about the transcripts you wish to provide. This can be a string or any object."
14082
- ),
14083
- answer_format: import_zod3.z.string().optional().describe(
14084
- `How you want the answer to be returned. This can be any text. Can't be used with answer_options. Examples: "short sentence", "bullet points"
14085
- `
14086
- ),
14087
- answer_options: import_zod3.z.array(import_zod3.z.string()).optional().describe(
14088
- `What discrete options to return. Useful for precise responses. Can't be used with answer_format. Example: ["Yes", "No"]
14089
- `
14090
- )
14091
- })
14092
- ).describe("A list of questions to ask")
14093
- })
14094
- );
14095
- var lemurQuestionAnswerResponseUsageInputTokensMin = 0;
14096
- var lemurQuestionAnswerResponseUsageOutputTokensMin = 0;
14097
- var lemurQuestionAnswerResponse = import_zod3.z.object({
14098
- request_id: import_zod3.z.string().uuid().describe("The ID of the LeMUR request"),
14099
- usage: import_zod3.z.object({
14100
- input_tokens: import_zod3.z.number().min(lemurQuestionAnswerResponseUsageInputTokensMin).describe("The number of input tokens used by the model"),
14101
- output_tokens: import_zod3.z.number().min(lemurQuestionAnswerResponseUsageOutputTokensMin).describe("The number of output tokens generated by the model")
14102
- }).describe("The usage numbers for the LeMUR request")
14103
- }).and(
14104
- import_zod3.z.object({
14105
- response: import_zod3.z.array(
14106
- import_zod3.z.object({
14107
- question: import_zod3.z.string().describe("The question for LeMUR to answer"),
14108
- answer: import_zod3.z.string().describe("The answer generated by LeMUR")
14109
- }).describe("An answer generated by LeMUR and its question")
14110
- ).describe("The answers generated by LeMUR and their questions")
14111
- })
14112
- );
14113
- var getLemurResponseParams = import_zod3.z.object({
14114
- request_id: import_zod3.z.string().describe(
14115
- "The ID of the LeMUR request you previously made.\nThis would be found in the response of the original request.\n"
14116
- )
14117
- });
14118
- var getLemurResponseResponseUsageInputTokensMin = 0;
14119
- var getLemurResponseResponseUsageOutputTokensMin = 0;
14120
- var getLemurResponseResponseUsageInputTokensMinOne = 0;
14121
- var getLemurResponseResponseUsageOutputTokensMinOne = 0;
14122
- var getLemurResponseResponse = import_zod3.z.object({
14123
- response: import_zod3.z.string().describe("The response generated by LeMUR.")
14124
- }).and(
14125
- import_zod3.z.object({
14126
- request_id: import_zod3.z.string().uuid().describe("The ID of the LeMUR request"),
14127
- usage: import_zod3.z.object({
14128
- input_tokens: import_zod3.z.number().min(getLemurResponseResponseUsageInputTokensMin).describe("The number of input tokens used by the model"),
14129
- output_tokens: import_zod3.z.number().min(getLemurResponseResponseUsageOutputTokensMin).describe("The number of output tokens generated by the model")
14130
- }).describe("The usage numbers for the LeMUR request")
14131
- })
14132
- ).or(
14133
- import_zod3.z.object({
14134
- request_id: import_zod3.z.string().uuid().describe("The ID of the LeMUR request"),
14135
- usage: import_zod3.z.object({
14136
- input_tokens: import_zod3.z.number().min(getLemurResponseResponseUsageInputTokensMinOne).describe("The number of input tokens used by the model"),
14137
- output_tokens: import_zod3.z.number().min(getLemurResponseResponseUsageOutputTokensMinOne).describe("The number of output tokens generated by the model")
14138
- }).describe("The usage numbers for the LeMUR request")
14139
- }).and(
14140
- import_zod3.z.object({
14141
- response: import_zod3.z.array(
14142
- import_zod3.z.object({
14143
- question: import_zod3.z.string().describe("The question for LeMUR to answer"),
14144
- answer: import_zod3.z.string().describe("The answer generated by LeMUR")
14145
- }).describe("An answer generated by LeMUR and its question")
14146
- ).describe("The answers generated by LeMUR and their questions")
14147
- })
14148
- )
14149
- );
14150
- var purgeLemurRequestDataParams = import_zod3.z.object({
14151
- request_id: import_zod3.z.string().describe(
14152
- "The ID of the LeMUR request whose data you want to delete. This would be found in the response of the original request."
14153
- )
14154
- });
14155
- var purgeLemurRequestDataResponse = import_zod3.z.object({
14156
- request_id: import_zod3.z.string().uuid().describe("The ID of the deletion request of the LeMUR request"),
14157
- request_id_to_purge: import_zod3.z.string().uuid().describe("The ID of the LeMUR request to purge the data for"),
14158
- deleted: import_zod3.z.boolean().describe("Whether the request data was deleted")
14159
- });
14160
14856
 
14161
14857
  // src/generated/assemblyai/streaming-types.zod.ts
14162
14858
  var import_zod4 = require("zod");
@@ -14171,25 +14867,37 @@ var streamingTranscriberParams = import_zod4.z.object({
14171
14867
  enableExtraSessionInformation: import_zod4.z.boolean().optional().describe(
14172
14868
  "Set to true to receive the SessionInformation message before the session ends. Defaults to false."
14173
14869
  ),
14870
+ domain: import_zod4.z.string().optional().describe(
14871
+ 'Enable domain-specific transcription models to improve accuracy for specialized terminology. Set to `"medical-v1"` to enable [Medical Mode](https://www.assemblyai.com/docs/streaming/medical-mode) for improved accuracy of medical terms such as medications, procedures, conditions, and dosages. Supported languages: English (`en`), Spanish (`es`), German (`de`), French (`fr`). If used with an unsupported language, the parameter is ignored and a warning is returned.'
14872
+ ),
14174
14873
  endOfTurnConfidenceThreshold: import_zod4.z.number().optional().describe("From SDK v3"),
14175
14874
  minEndOfTurnSilenceWhenConfident: import_zod4.z.number().optional().describe("From SDK v3"),
14875
+ minTurnSilence: import_zod4.z.number().optional().describe("From SDK v3"),
14176
14876
  maxTurnSilence: import_zod4.z.number().optional().describe("From SDK v3"),
14177
14877
  vadThreshold: import_zod4.z.number().optional().describe("From SDK v3"),
14178
14878
  formatTurns: import_zod4.z.boolean().optional().describe("From SDK v3"),
14179
14879
  filterProfanity: import_zod4.z.boolean().optional().describe("From SDK v3"),
14180
14880
  keyterms: import_zod4.z.array(import_zod4.z.string()).optional().describe("From SDK v3"),
14181
14881
  keytermsPrompt: import_zod4.z.array(import_zod4.z.string()).optional().describe("From SDK v3"),
14182
- speechModel: import_zod4.z.enum(["universal-streaming-english", "universal-streaming-multilingual"]).optional().describe("From SDK v3"),
14882
+ prompt: import_zod4.z.string().optional().describe("From SDK v3"),
14883
+ speechModel: import_zod4.z.enum(["universal-streaming-english", "universal-streaming-multilingual"]).describe("From SDK v3"),
14183
14884
  languageDetection: import_zod4.z.boolean().optional().describe("From SDK v3"),
14184
- inactivityTimeout: import_zod4.z.number().optional().describe("From SDK v3")
14885
+ inactivityTimeout: import_zod4.z.number().optional().describe("From SDK v3"),
14886
+ speakerLabels: import_zod4.z.boolean().optional().describe("From SDK v3"),
14887
+ maxSpeakers: import_zod4.z.number().optional().describe("From SDK v3"),
14888
+ llmGateway: import_zod4.z.unknown().optional().describe("From SDK v3")
14185
14889
  });
14186
14890
  var streamingUpdateConfigParams = import_zod4.z.object({
14187
14891
  end_utterance_silence_threshold: import_zod4.z.number().min(0).max(2e4).optional().describe("The duration threshold in milliseconds"),
14188
14892
  end_of_turn_confidence_threshold: import_zod4.z.number().optional().describe("From SDK v3"),
14189
14893
  min_end_of_turn_silence_when_confident: import_zod4.z.number().optional().describe("From SDK v3"),
14894
+ min_turn_silence: import_zod4.z.number().optional().describe("From SDK v3"),
14190
14895
  max_turn_silence: import_zod4.z.number().optional().describe("From SDK v3"),
14191
14896
  vad_threshold: import_zod4.z.number().optional().describe("From SDK v3"),
14192
- format_turns: import_zod4.z.boolean().optional().describe("From SDK v3")
14897
+ format_turns: import_zod4.z.boolean().optional().describe("From SDK v3"),
14898
+ keyterms_prompt: import_zod4.z.array(import_zod4.z.string()).optional().describe("From SDK v3"),
14899
+ prompt: import_zod4.z.string().optional().describe("From SDK v3"),
14900
+ filter_profanity: import_zod4.z.boolean().optional().describe("From SDK v3")
14193
14901
  });
14194
14902
 
14195
14903
  // src/generated/gladia/api/gladiaControlAPI.zod.ts
@@ -32165,7 +32873,7 @@ var createRealtimeClientSecretBody = import_zod6.z.object({
32165
32873
  format: import_zod6.z.discriminatedUnion("type", [
32166
32874
  import_zod6.z.object({
32167
32875
  type: import_zod6.z.enum(["audio/pcm"]).describe("The audio format. Always `audio/pcm`."),
32168
- rate: import_zod6.z.literal(24e3).optional().describe("The sample rate of the audio. Always `24000`.")
32876
+ rate: import_zod6.z.literal(24e3).describe("The sample rate of the audio. Always `24000`.")
32169
32877
  }).describe("The PCM audio format. Only a 24kHz sample rate is supported."),
32170
32878
  import_zod6.z.object({
32171
32879
  type: import_zod6.z.enum(["audio/pcmu"]).describe("The audio format. Always `audio/pcmu`.")
@@ -32264,7 +32972,7 @@ var createRealtimeClientSecretBody = import_zod6.z.object({
32264
32972
  format: import_zod6.z.discriminatedUnion("type", [
32265
32973
  import_zod6.z.object({
32266
32974
  type: import_zod6.z.enum(["audio/pcm"]).describe("The audio format. Always `audio/pcm`."),
32267
- rate: import_zod6.z.literal(24e3).optional().describe("The sample rate of the audio. Always `24000`.")
32975
+ rate: import_zod6.z.literal(24e3).describe("The sample rate of the audio. Always `24000`.")
32268
32976
  }).describe("The PCM audio format. Only a 24kHz sample rate is supported."),
32269
32977
  import_zod6.z.object({
32270
32978
  type: import_zod6.z.enum(["audio/pcmu"]).describe("The audio format. Always `audio/pcmu`.")
@@ -32474,7 +33182,7 @@ var createRealtimeClientSecretBody = import_zod6.z.object({
32474
33182
  format: import_zod6.z.discriminatedUnion("type", [
32475
33183
  import_zod6.z.object({
32476
33184
  type: import_zod6.z.enum(["audio/pcm"]).describe("The audio format. Always `audio/pcm`."),
32477
- rate: import_zod6.z.literal(24e3).optional().describe("The sample rate of the audio. Always `24000`.")
33185
+ rate: import_zod6.z.literal(24e3).describe("The sample rate of the audio. Always `24000`.")
32478
33186
  }).describe("The PCM audio format. Only a 24kHz sample rate is supported."),
32479
33187
  import_zod6.z.object({
32480
33188
  type: import_zod6.z.enum(["audio/pcmu"]).describe("The audio format. Always `audio/pcmu`.")
@@ -32644,7 +33352,7 @@ var createRealtimeClientSecretResponse = import_zod6.z.object({
32644
33352
  format: import_zod6.z.discriminatedUnion("type", [
32645
33353
  import_zod6.z.object({
32646
33354
  type: import_zod6.z.enum(["audio/pcm"]).describe("The audio format. Always `audio/pcm`."),
32647
- rate: import_zod6.z.literal(24e3).optional().describe("The sample rate of the audio. Always `24000`.")
33355
+ rate: import_zod6.z.literal(24e3).describe("The sample rate of the audio. Always `24000`.")
32648
33356
  }).describe("The PCM audio format. Only a 24kHz sample rate is supported."),
32649
33357
  import_zod6.z.object({
32650
33358
  type: import_zod6.z.enum(["audio/pcmu"]).describe("The audio format. Always `audio/pcmu`.")
@@ -32743,7 +33451,7 @@ var createRealtimeClientSecretResponse = import_zod6.z.object({
32743
33451
  format: import_zod6.z.discriminatedUnion("type", [
32744
33452
  import_zod6.z.object({
32745
33453
  type: import_zod6.z.enum(["audio/pcm"]).describe("The audio format. Always `audio/pcm`."),
32746
- rate: import_zod6.z.literal(24e3).optional().describe("The sample rate of the audio. Always `24000`.")
33454
+ rate: import_zod6.z.literal(24e3).describe("The sample rate of the audio. Always `24000`.")
32747
33455
  }).describe("The PCM audio format. Only a 24kHz sample rate is supported."),
32748
33456
  import_zod6.z.object({
32749
33457
  type: import_zod6.z.enum(["audio/pcmu"]).describe("The audio format. Always `audio/pcmu`.")
@@ -32962,7 +33670,7 @@ var createRealtimeClientSecretResponse = import_zod6.z.object({
32962
33670
  format: import_zod6.z.discriminatedUnion("type", [
32963
33671
  import_zod6.z.object({
32964
33672
  type: import_zod6.z.enum(["audio/pcm"]).describe("The audio format. Always `audio/pcm`."),
32965
- rate: import_zod6.z.literal(24e3).optional().describe("The sample rate of the audio. Always `24000`.")
33673
+ rate: import_zod6.z.literal(24e3).describe("The sample rate of the audio. Always `24000`.")
32966
33674
  }).describe("The PCM audio format. Only a 24kHz sample rate is supported."),
32967
33675
  import_zod6.z.object({
32968
33676
  type: import_zod6.z.enum(["audio/pcmu"]).describe("The audio format. Always `audio/pcmu`.")
@@ -33191,7 +33899,7 @@ var createRealtimeSessionResponse = import_zod6.z.object({
33191
33899
  format: import_zod6.z.discriminatedUnion("type", [
33192
33900
  import_zod6.z.object({
33193
33901
  type: import_zod6.z.enum(["audio/pcm"]).describe("The audio format. Always `audio/pcm`."),
33194
- rate: import_zod6.z.literal(24e3).optional().describe("The sample rate of the audio. Always `24000`.")
33902
+ rate: import_zod6.z.literal(24e3).describe("The sample rate of the audio. Always `24000`.")
33195
33903
  }).describe("The PCM audio format. Only a 24kHz sample rate is supported."),
33196
33904
  import_zod6.z.object({
33197
33905
  type: import_zod6.z.enum(["audio/pcmu"]).describe("The audio format. Always `audio/pcmu`.")
@@ -33235,7 +33943,7 @@ var createRealtimeSessionResponse = import_zod6.z.object({
33235
33943
  format: import_zod6.z.discriminatedUnion("type", [
33236
33944
  import_zod6.z.object({
33237
33945
  type: import_zod6.z.enum(["audio/pcm"]).describe("The audio format. Always `audio/pcm`."),
33238
- rate: import_zod6.z.literal(24e3).optional().describe("The sample rate of the audio. Always `24000`.")
33946
+ rate: import_zod6.z.literal(24e3).describe("The sample rate of the audio. Always `24000`.")
33239
33947
  }).describe("The PCM audio format. Only a 24kHz sample rate is supported."),
33240
33948
  import_zod6.z.object({
33241
33949
  type: import_zod6.z.enum(["audio/pcmu"]).describe("The audio format. Always `audio/pcmu`.")
@@ -38333,20 +39041,6 @@ var LanguagePackInfoWritingDirection = {
38333
39041
  "right-to-left": "right-to-left"
38334
39042
  };
38335
39043
 
38336
- // src/generated/speechmatics/schema/notificationConfigContentsItem.ts
38337
- var NotificationConfigContentsItem = {
38338
- jobinfo: "jobinfo",
38339
- transcript: "transcript",
38340
- "transcriptjson-v2": "transcript.json-v2",
38341
- transcripttxt: "transcript.txt",
38342
- transcriptsrt: "transcript.srt",
38343
- alignment: "alignment",
38344
- alignmentword_start_and_end: "alignment.word_start_and_end",
38345
- alignmentone_per_line: "alignment.one_per_line",
38346
- data: "data",
38347
- text: "text"
38348
- };
38349
-
38350
39044
  // src/generated/speechmatics/schema/notificationConfigMethod.ts
38351
39045
  var NotificationConfigMethod = {
38352
39046
  post: "post",
@@ -40227,7 +40921,6 @@ var deleteTranscriptByIdResponse = import_zod13.z.any();
40227
40921
  createOpenAIWhisperAdapter,
40228
40922
  createSonioxAdapter,
40229
40923
  createSpeechmaticsAdapter,
40230
- createTemporaryToken,
40231
40924
  createTranscript,
40232
40925
  createTranscription,
40233
40926
  createVoiceRouter,
@@ -40279,6 +40972,9 @@ var deleteTranscriptByIdResponse = import_zod13.z.any();
40279
40972
  transcriptionsGet,
40280
40973
  transcriptionsList,
40281
40974
  transcriptionsListFiles,
40975
+ webHooksCreate,
40976
+ webHooksDelete,
40977
+ webHooksList,
40282
40978
  zodToFieldConfigs
40283
40979
  });
40284
40980
  //# sourceMappingURL=index.js.map