voice-router-dev 0.8.3 → 0.8.5

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,
@@ -2482,13 +2481,6 @@ var TranslationLanguageCodeEnum = {
2482
2481
  zh: "zh"
2483
2482
  };
2484
2483
 
2485
- // src/generated/assemblyai/schema/speechModel.ts
2486
- var SpeechModel = {
2487
- best: "best",
2488
- "slam-1": "slam-1",
2489
- universal: "universal"
2490
- };
2491
-
2492
2484
  // src/generated/assemblyai/schema/transcriptLanguageCode.ts
2493
2485
  var TranscriptLanguageCode = {
2494
2486
  en: "en",
@@ -2804,7 +2796,10 @@ var AssemblyAIEncoding = {
2804
2796
  /** μ-law (telephony) */
2805
2797
  pcmMulaw: "pcm_mulaw"
2806
2798
  };
2807
- var AssemblyAITranscriptionModel = SpeechModel;
2799
+ var AssemblyAITranscriptionModel = {
2800
+ "universal-3-pro": "universal-3-pro",
2801
+ "universal-2": "universal-2"
2802
+ };
2808
2803
  var AssemblyAILanguage = TranscriptLanguageCode;
2809
2804
  var AssemblyAISpeechModel = {
2810
2805
  /** Optimized for English */
@@ -5429,19 +5424,20 @@ var schema_exports2 = {};
5429
5424
  __export(schema_exports2, {
5430
5425
  AudioIntelligenceModelStatus: () => AudioIntelligenceModelStatus,
5431
5426
  EntityType: () => EntityType,
5432
- LemurModel: () => LemurModel,
5433
5427
  PiiPolicy: () => PiiPolicy,
5434
5428
  RedactPiiAudioQuality: () => RedactPiiAudioQuality,
5435
5429
  RedactedAudioStatus: () => RedactedAudioStatus,
5436
5430
  Sentiment: () => Sentiment,
5437
- SpeechModel: () => SpeechModel,
5431
+ SpeakerIdentificationRequestBodySpeakerIdentificationSpeakerType: () => SpeakerIdentificationRequestBodySpeakerIdentificationSpeakerType,
5438
5432
  SubstitutionPolicy: () => SubstitutionPolicy,
5439
5433
  SubtitleFormat: () => SubtitleFormat,
5440
5434
  SummaryModel: () => SummaryModel,
5441
5435
  SummaryType: () => SummaryType,
5442
- TranscriptBoostParam: () => TranscriptBoostParam,
5443
5436
  TranscriptLanguageCode: () => TranscriptLanguageCode,
5444
- TranscriptReadyStatus: () => TranscriptReadyStatus,
5437
+ TranscriptOptionalParamsRedactPiiAudioOptionsOverrideAudioRedactionMethod: () => TranscriptOptionalParamsRedactPiiAudioOptionsOverrideAudioRedactionMethod,
5438
+ TranscriptOptionalParamsRemoveAudioTags: () => TranscriptOptionalParamsRemoveAudioTags,
5439
+ TranscriptRedactPiiAudioOptionsOverrideAudioRedactionMethod: () => TranscriptRedactPiiAudioOptionsOverrideAudioRedactionMethod,
5440
+ TranscriptRemoveAudioTags: () => TranscriptRemoveAudioTags,
5445
5441
  TranscriptStatus: () => TranscriptStatus
5446
5442
  });
5447
5443
 
@@ -5499,13 +5495,6 @@ var EntityType = {
5499
5495
  zodiac_sign: "zodiac_sign"
5500
5496
  };
5501
5497
 
5502
- // src/generated/assemblyai/schema/lemurModel.ts
5503
- var LemurModel = {
5504
- "anthropic/claude-3-5-sonnet": "anthropic/claude-3-5-sonnet",
5505
- "anthropic/claude-3-opus": "anthropic/claude-3-opus",
5506
- "anthropic/claude-3-haiku": "anthropic/claude-3-haiku"
5507
- };
5508
-
5509
5498
  // src/generated/assemblyai/schema/piiPolicy.ts
5510
5499
  var PiiPolicy = {
5511
5500
  account_number: "account_number",
@@ -5572,6 +5561,12 @@ var Sentiment = {
5572
5561
  NEGATIVE: "NEGATIVE"
5573
5562
  };
5574
5563
 
5564
+ // src/generated/assemblyai/schema/speakerIdentificationRequestBodySpeakerIdentificationSpeakerType.ts
5565
+ var SpeakerIdentificationRequestBodySpeakerIdentificationSpeakerType = {
5566
+ role: "role",
5567
+ name: "name"
5568
+ };
5569
+
5575
5570
  // src/generated/assemblyai/schema/substitutionPolicy.ts
5576
5571
  var SubstitutionPolicy = {
5577
5572
  entity_name: "entity_name",
@@ -5600,17 +5595,24 @@ var SummaryType = {
5600
5595
  paragraph: "paragraph"
5601
5596
  };
5602
5597
 
5603
- // src/generated/assemblyai/schema/transcriptBoostParam.ts
5604
- var TranscriptBoostParam = {
5605
- low: "low",
5606
- default: "default",
5607
- high: "high"
5598
+ // src/generated/assemblyai/schema/transcriptOptionalParamsRedactPiiAudioOptionsOverrideAudioRedactionMethod.ts
5599
+ var TranscriptOptionalParamsRedactPiiAudioOptionsOverrideAudioRedactionMethod = {
5600
+ silence: "silence"
5608
5601
  };
5609
5602
 
5610
- // src/generated/assemblyai/schema/transcriptReadyStatus.ts
5611
- var TranscriptReadyStatus = {
5612
- completed: "completed",
5613
- 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"
5614
5616
  };
5615
5617
 
5616
5618
  // src/generated/assemblyai/api/assemblyAIAPI.ts
@@ -5629,9 +5631,6 @@ var getTranscript = (transcriptId, options) => {
5629
5631
  var deleteTranscript = (transcriptId, options) => {
5630
5632
  return import_axios2.default.delete(`/v2/transcript/${transcriptId}`, options);
5631
5633
  };
5632
- var createTemporaryToken = (createRealtimeTemporaryTokenParams, options) => {
5633
- return import_axios2.default.post(`/v2/realtime/token`, createRealtimeTemporaryTokenParams, options);
5634
- };
5635
5634
 
5636
5635
  // src/adapters/assemblyai-adapter.ts
5637
5636
  var AssemblyAIAdapter = class extends BaseAdapter {
@@ -5967,16 +5966,27 @@ var AssemblyAIAdapter = class extends BaseAdapter {
5967
5966
  "AssemblyAI adapter currently only supports URL-based audio input. Use audio.type='url'"
5968
5967
  );
5969
5968
  }
5969
+ const aaiOpts = { ...options?.assemblyai };
5970
+ if ("speech_model" in aaiOpts && aaiOpts.speech_model != null) {
5971
+ if (!aaiOpts.speech_models) {
5972
+ aaiOpts.speech_models = [aaiOpts.speech_model];
5973
+ }
5974
+ delete aaiOpts.speech_model;
5975
+ }
5970
5976
  const request = {
5971
- ...options?.assemblyai,
5977
+ ...aaiOpts,
5972
5978
  audio_url: audioUrl,
5979
+ // speech_models is required — default to universal-3-pro
5980
+ speech_models: aaiOpts.speech_models ?? [
5981
+ "universal-3-pro"
5982
+ ],
5973
5983
  // Enable punctuation and formatting by default
5974
- punctuate: options?.assemblyai?.punctuate ?? true,
5975
- format_text: options?.assemblyai?.format_text ?? true
5984
+ punctuate: aaiOpts.punctuate ?? true,
5985
+ format_text: aaiOpts.format_text ?? true
5976
5986
  };
5977
5987
  if (options) {
5978
5988
  if (options.model) {
5979
- request.speech_model = options.model;
5989
+ request.speech_models = [options.model];
5980
5990
  }
5981
5991
  if (options.language) {
5982
5992
  const languageCode = options.language.includes("_") ? options.language : `${options.language}_us`;
@@ -5992,8 +6002,7 @@ var AssemblyAIAdapter = class extends BaseAdapter {
5992
6002
  }
5993
6003
  }
5994
6004
  if (options.customVocabulary && options.customVocabulary.length > 0) {
5995
- request.word_boost = options.customVocabulary;
5996
- request.boost_param = request.boost_param ?? "high";
6005
+ request.keyterms_prompt = options.customVocabulary;
5997
6006
  }
5998
6007
  if (options.summarization) {
5999
6008
  request.summarization = true;
@@ -11773,15 +11782,10 @@ var deepgramStreamingOnlyParams = import_zod2.z.object({
11773
11782
  // src/generated/assemblyai/api/assemblyAIAPI.zod.ts
11774
11783
  var assemblyAIAPI_zod_exports = {};
11775
11784
  __export(assemblyAIAPI_zod_exports, {
11776
- createTemporaryTokenBody: () => createTemporaryTokenBody,
11777
- createTemporaryTokenBodyExpiresInMin: () => createTemporaryTokenBodyExpiresInMin,
11778
- createTemporaryTokenResponse: () => createTemporaryTokenResponse,
11779
11785
  createTranscriptBody: () => createTranscriptBody,
11780
11786
  createTranscriptBodyAutoChaptersDefault: () => createTranscriptBodyAutoChaptersDefault,
11781
11787
  createTranscriptBodyAutoHighlightsDefault: () => createTranscriptBodyAutoHighlightsDefault,
11782
11788
  createTranscriptBodyContentSafetyConfidenceDefault: () => createTranscriptBodyContentSafetyConfidenceDefault,
11783
- createTranscriptBodyContentSafetyConfidenceMax: () => createTranscriptBodyContentSafetyConfidenceMax,
11784
- createTranscriptBodyContentSafetyConfidenceMin: () => createTranscriptBodyContentSafetyConfidenceMin,
11785
11789
  createTranscriptBodyContentSafetyDefault: () => createTranscriptBodyContentSafetyDefault,
11786
11790
  createTranscriptBodyCustomTopicsDefault: () => createTranscriptBodyCustomTopicsDefault,
11787
11791
  createTranscriptBodyDisfluenciesDefault: () => createTranscriptBodyDisfluenciesDefault,
@@ -11789,198 +11793,58 @@ __export(assemblyAIAPI_zod_exports, {
11789
11793
  createTranscriptBodyFilterProfanityDefault: () => createTranscriptBodyFilterProfanityDefault,
11790
11794
  createTranscriptBodyFormatTextDefault: () => createTranscriptBodyFormatTextDefault,
11791
11795
  createTranscriptBodyIabCategoriesDefault: () => createTranscriptBodyIabCategoriesDefault,
11792
- createTranscriptBodyLanguageCodeDefault: () => createTranscriptBodyLanguageCodeDefault,
11793
- createTranscriptBodyLanguageConfidenceThresholdDefault: () => createTranscriptBodyLanguageConfidenceThresholdDefault,
11794
- createTranscriptBodyLanguageConfidenceThresholdMax: () => createTranscriptBodyLanguageConfidenceThresholdMax,
11795
- createTranscriptBodyLanguageConfidenceThresholdMin: () => createTranscriptBodyLanguageConfidenceThresholdMin,
11796
11796
  createTranscriptBodyLanguageDetectionDefault: () => createTranscriptBodyLanguageDetectionDefault,
11797
+ createTranscriptBodyLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault: () => createTranscriptBodyLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault,
11798
+ createTranscriptBodyLanguageDetectionOptionsCodeSwitchingDefault: () => createTranscriptBodyLanguageDetectionOptionsCodeSwitchingDefault,
11799
+ createTranscriptBodyLanguageDetectionOptionsFallbackLanguageDefault: () => createTranscriptBodyLanguageDetectionOptionsFallbackLanguageDefault,
11797
11800
  createTranscriptBodyMultichannelDefault: () => createTranscriptBodyMultichannelDefault,
11798
11801
  createTranscriptBodyPunctuateDefault: () => createTranscriptBodyPunctuateDefault,
11799
11802
  createTranscriptBodyRedactPiiAudioDefault: () => createTranscriptBodyRedactPiiAudioDefault,
11803
+ createTranscriptBodyRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault: () => createTranscriptBodyRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault,
11800
11804
  createTranscriptBodyRedactPiiDefault: () => createTranscriptBodyRedactPiiDefault,
11801
- createTranscriptBodyRedactPiiSubDefault: () => createTranscriptBodyRedactPiiSubDefault,
11802
11805
  createTranscriptBodySentimentAnalysisDefault: () => createTranscriptBodySentimentAnalysisDefault,
11803
11806
  createTranscriptBodySpeakerLabelsDefault: () => createTranscriptBodySpeakerLabelsDefault,
11804
- createTranscriptBodySpeakersExpectedDefault: () => createTranscriptBodySpeakersExpectedDefault,
11805
- createTranscriptBodySpeechModelDefault: () => createTranscriptBodySpeechModelDefault,
11806
- createTranscriptBodySpeechThresholdDefault: () => createTranscriptBodySpeechThresholdDefault,
11807
- createTranscriptBodySpeechThresholdMax: () => createTranscriptBodySpeechThresholdMax,
11808
- createTranscriptBodySpeechThresholdMin: () => createTranscriptBodySpeechThresholdMin,
11807
+ createTranscriptBodySpeakerOptionsMinSpeakersExpectedDefault: () => createTranscriptBodySpeakerOptionsMinSpeakersExpectedDefault,
11808
+ createTranscriptBodySpeechUnderstandingRequestTranslationFormalDefault: () => createTranscriptBodySpeechUnderstandingRequestTranslationFormalDefault,
11809
+ createTranscriptBodySpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault: () => createTranscriptBodySpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault,
11809
11810
  createTranscriptBodySummarizationDefault: () => createTranscriptBodySummarizationDefault,
11810
- createTranscriptBodyWebhookAuthHeaderNameDefault: () => createTranscriptBodyWebhookAuthHeaderNameDefault,
11811
- createTranscriptBodyWebhookAuthHeaderValueDefault: () => createTranscriptBodyWebhookAuthHeaderValueDefault,
11811
+ createTranscriptBodyTemperatureDefault: () => createTranscriptBodyTemperatureDefault,
11812
11812
  createTranscriptResponse: () => createTranscriptResponse,
11813
- createTranscriptResponseAutoHighlightsResultResultsItemRankMax: () => createTranscriptResponseAutoHighlightsResultResultsItemRankMax,
11814
- createTranscriptResponseAutoHighlightsResultResultsItemRankMin: () => createTranscriptResponseAutoHighlightsResultResultsItemRankMin,
11815
- createTranscriptResponseConfidenceMax: () => createTranscriptResponseConfidenceMax,
11816
- createTranscriptResponseConfidenceMin: () => createTranscriptResponseConfidenceMin,
11817
- createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax: () => createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax,
11818
- createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin: () => createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin,
11819
- createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax: () => createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax,
11820
- createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin: () => createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin,
11821
- createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax: () => createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax,
11822
- createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin: () => createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin,
11823
- createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax: () => createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax,
11824
- createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin: () => createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin,
11825
- createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax: () => createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax,
11826
- createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin: () => createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin,
11827
- createTranscriptResponseContentSafetyLabelsSummaryMaxOne: () => createTranscriptResponseContentSafetyLabelsSummaryMaxOne,
11828
- createTranscriptResponseContentSafetyLabelsSummaryMinOne: () => createTranscriptResponseContentSafetyLabelsSummaryMinOne,
11829
- createTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax: () => createTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax,
11830
- createTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin: () => createTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin,
11831
- createTranscriptResponseIabCategoriesResultSummaryMaxOne: () => createTranscriptResponseIabCategoriesResultSummaryMaxOne,
11832
- createTranscriptResponseIabCategoriesResultSummaryMinOne: () => createTranscriptResponseIabCategoriesResultSummaryMinOne,
11833
- createTranscriptResponseLanguageConfidenceMax: () => createTranscriptResponseLanguageConfidenceMax,
11834
- createTranscriptResponseLanguageConfidenceMin: () => createTranscriptResponseLanguageConfidenceMin,
11835
- createTranscriptResponseLanguageConfidenceThresholdMax: () => createTranscriptResponseLanguageConfidenceThresholdMax,
11836
- createTranscriptResponseLanguageConfidenceThresholdMin: () => createTranscriptResponseLanguageConfidenceThresholdMin,
11837
- createTranscriptResponseSentimentAnalysisResultsItemConfidenceMax: () => createTranscriptResponseSentimentAnalysisResultsItemConfidenceMax,
11838
- createTranscriptResponseSentimentAnalysisResultsItemConfidenceMin: () => createTranscriptResponseSentimentAnalysisResultsItemConfidenceMin,
11839
- createTranscriptResponseSpeechModelDefault: () => createTranscriptResponseSpeechModelDefault,
11840
- createTranscriptResponseSpeechThresholdMax: () => createTranscriptResponseSpeechThresholdMax,
11841
- createTranscriptResponseSpeechThresholdMin: () => createTranscriptResponseSpeechThresholdMin,
11842
- createTranscriptResponseUtterancesItemConfidenceMax: () => createTranscriptResponseUtterancesItemConfidenceMax,
11843
- createTranscriptResponseUtterancesItemConfidenceMin: () => createTranscriptResponseUtterancesItemConfidenceMin,
11844
- createTranscriptResponseUtterancesItemWordsItemConfidenceMax: () => createTranscriptResponseUtterancesItemWordsItemConfidenceMax,
11845
- createTranscriptResponseUtterancesItemWordsItemConfidenceMin: () => createTranscriptResponseUtterancesItemWordsItemConfidenceMin,
11846
- createTranscriptResponseWordsItemConfidenceMax: () => createTranscriptResponseWordsItemConfidenceMax,
11847
- createTranscriptResponseWordsItemConfidenceMin: () => createTranscriptResponseWordsItemConfidenceMin,
11813
+ createTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault: () => createTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault,
11814
+ createTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault: () => createTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault,
11815
+ createTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault: () => createTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault,
11816
+ createTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault: () => createTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault,
11817
+ createTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault: () => createTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault,
11818
+ createTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault: () => createTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault,
11848
11819
  deleteTranscriptParams: () => deleteTranscriptParams,
11849
11820
  deleteTranscriptResponse: () => deleteTranscriptResponse,
11850
- deleteTranscriptResponseAutoHighlightsResultResultsItemRankMax: () => deleteTranscriptResponseAutoHighlightsResultResultsItemRankMax,
11851
- deleteTranscriptResponseAutoHighlightsResultResultsItemRankMin: () => deleteTranscriptResponseAutoHighlightsResultResultsItemRankMin,
11852
- deleteTranscriptResponseConfidenceMax: () => deleteTranscriptResponseConfidenceMax,
11853
- deleteTranscriptResponseConfidenceMin: () => deleteTranscriptResponseConfidenceMin,
11854
- deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax: () => deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax,
11855
- deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin: () => deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin,
11856
- deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax: () => deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax,
11857
- deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin: () => deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin,
11858
- deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax: () => deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax,
11859
- deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin: () => deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin,
11860
- deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax: () => deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax,
11861
- deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin: () => deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin,
11862
- deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax: () => deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax,
11863
- deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin: () => deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin,
11864
- deleteTranscriptResponseContentSafetyLabelsSummaryMaxOne: () => deleteTranscriptResponseContentSafetyLabelsSummaryMaxOne,
11865
- deleteTranscriptResponseContentSafetyLabelsSummaryMinOne: () => deleteTranscriptResponseContentSafetyLabelsSummaryMinOne,
11866
- deleteTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax: () => deleteTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax,
11867
- deleteTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin: () => deleteTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin,
11868
- deleteTranscriptResponseIabCategoriesResultSummaryMaxOne: () => deleteTranscriptResponseIabCategoriesResultSummaryMaxOne,
11869
- deleteTranscriptResponseIabCategoriesResultSummaryMinOne: () => deleteTranscriptResponseIabCategoriesResultSummaryMinOne,
11870
- deleteTranscriptResponseLanguageConfidenceMax: () => deleteTranscriptResponseLanguageConfidenceMax,
11871
- deleteTranscriptResponseLanguageConfidenceMin: () => deleteTranscriptResponseLanguageConfidenceMin,
11872
- deleteTranscriptResponseLanguageConfidenceThresholdMax: () => deleteTranscriptResponseLanguageConfidenceThresholdMax,
11873
- deleteTranscriptResponseLanguageConfidenceThresholdMin: () => deleteTranscriptResponseLanguageConfidenceThresholdMin,
11874
- deleteTranscriptResponseSentimentAnalysisResultsItemConfidenceMax: () => deleteTranscriptResponseSentimentAnalysisResultsItemConfidenceMax,
11875
- deleteTranscriptResponseSentimentAnalysisResultsItemConfidenceMin: () => deleteTranscriptResponseSentimentAnalysisResultsItemConfidenceMin,
11876
- deleteTranscriptResponseSpeechModelDefault: () => deleteTranscriptResponseSpeechModelDefault,
11877
- deleteTranscriptResponseSpeechThresholdMax: () => deleteTranscriptResponseSpeechThresholdMax,
11878
- deleteTranscriptResponseSpeechThresholdMin: () => deleteTranscriptResponseSpeechThresholdMin,
11879
- deleteTranscriptResponseUtterancesItemConfidenceMax: () => deleteTranscriptResponseUtterancesItemConfidenceMax,
11880
- deleteTranscriptResponseUtterancesItemConfidenceMin: () => deleteTranscriptResponseUtterancesItemConfidenceMin,
11881
- deleteTranscriptResponseUtterancesItemWordsItemConfidenceMax: () => deleteTranscriptResponseUtterancesItemWordsItemConfidenceMax,
11882
- deleteTranscriptResponseUtterancesItemWordsItemConfidenceMin: () => deleteTranscriptResponseUtterancesItemWordsItemConfidenceMin,
11883
- deleteTranscriptResponseWordsItemConfidenceMax: () => deleteTranscriptResponseWordsItemConfidenceMax,
11884
- deleteTranscriptResponseWordsItemConfidenceMin: () => deleteTranscriptResponseWordsItemConfidenceMin,
11885
- getLemurResponseParams: () => getLemurResponseParams,
11886
- getLemurResponseResponse: () => getLemurResponseResponse,
11887
- getLemurResponseResponseUsageInputTokensMin: () => getLemurResponseResponseUsageInputTokensMin,
11888
- getLemurResponseResponseUsageInputTokensMinOne: () => getLemurResponseResponseUsageInputTokensMinOne,
11889
- getLemurResponseResponseUsageOutputTokensMin: () => getLemurResponseResponseUsageOutputTokensMin,
11890
- getLemurResponseResponseUsageOutputTokensMinOne: () => getLemurResponseResponseUsageOutputTokensMinOne,
11821
+ deleteTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault: () => deleteTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault,
11822
+ deleteTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault: () => deleteTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault,
11823
+ deleteTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault: () => deleteTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault,
11824
+ deleteTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault: () => deleteTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault,
11825
+ deleteTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault: () => deleteTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault,
11826
+ deleteTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault: () => deleteTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault,
11891
11827
  getRedactedAudioParams: () => getRedactedAudioParams,
11892
11828
  getRedactedAudioResponse: () => getRedactedAudioResponse,
11893
11829
  getSubtitlesParams: () => getSubtitlesParams,
11894
11830
  getSubtitlesQueryParams: () => getSubtitlesQueryParams,
11831
+ getSubtitlesResponse: () => getSubtitlesResponse,
11895
11832
  getTranscriptParagraphsParams: () => getTranscriptParagraphsParams,
11896
11833
  getTranscriptParagraphsResponse: () => getTranscriptParagraphsResponse,
11897
- getTranscriptParagraphsResponseConfidenceMax: () => getTranscriptParagraphsResponseConfidenceMax,
11898
- getTranscriptParagraphsResponseConfidenceMin: () => getTranscriptParagraphsResponseConfidenceMin,
11899
- getTranscriptParagraphsResponseParagraphsItemConfidenceMax: () => getTranscriptParagraphsResponseParagraphsItemConfidenceMax,
11900
- getTranscriptParagraphsResponseParagraphsItemConfidenceMin: () => getTranscriptParagraphsResponseParagraphsItemConfidenceMin,
11901
- getTranscriptParagraphsResponseParagraphsItemWordsItemConfidenceMax: () => getTranscriptParagraphsResponseParagraphsItemWordsItemConfidenceMax,
11902
- getTranscriptParagraphsResponseParagraphsItemWordsItemConfidenceMin: () => getTranscriptParagraphsResponseParagraphsItemWordsItemConfidenceMin,
11903
11834
  getTranscriptParams: () => getTranscriptParams,
11904
11835
  getTranscriptResponse: () => getTranscriptResponse,
11905
- getTranscriptResponseAutoHighlightsResultResultsItemRankMax: () => getTranscriptResponseAutoHighlightsResultResultsItemRankMax,
11906
- getTranscriptResponseAutoHighlightsResultResultsItemRankMin: () => getTranscriptResponseAutoHighlightsResultResultsItemRankMin,
11907
- getTranscriptResponseConfidenceMax: () => getTranscriptResponseConfidenceMax,
11908
- getTranscriptResponseConfidenceMin: () => getTranscriptResponseConfidenceMin,
11909
- getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax: () => getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax,
11910
- getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin: () => getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin,
11911
- getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax: () => getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax,
11912
- getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin: () => getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin,
11913
- getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax: () => getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax,
11914
- getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin: () => getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin,
11915
- getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax: () => getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax,
11916
- getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin: () => getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin,
11917
- getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax: () => getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax,
11918
- getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin: () => getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin,
11919
- getTranscriptResponseContentSafetyLabelsSummaryMaxOne: () => getTranscriptResponseContentSafetyLabelsSummaryMaxOne,
11920
- getTranscriptResponseContentSafetyLabelsSummaryMinOne: () => getTranscriptResponseContentSafetyLabelsSummaryMinOne,
11921
- getTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax: () => getTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax,
11922
- getTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin: () => getTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin,
11923
- getTranscriptResponseIabCategoriesResultSummaryMaxOne: () => getTranscriptResponseIabCategoriesResultSummaryMaxOne,
11924
- getTranscriptResponseIabCategoriesResultSummaryMinOne: () => getTranscriptResponseIabCategoriesResultSummaryMinOne,
11925
- getTranscriptResponseLanguageConfidenceMax: () => getTranscriptResponseLanguageConfidenceMax,
11926
- getTranscriptResponseLanguageConfidenceMin: () => getTranscriptResponseLanguageConfidenceMin,
11927
- getTranscriptResponseLanguageConfidenceThresholdMax: () => getTranscriptResponseLanguageConfidenceThresholdMax,
11928
- getTranscriptResponseLanguageConfidenceThresholdMin: () => getTranscriptResponseLanguageConfidenceThresholdMin,
11929
- getTranscriptResponseSentimentAnalysisResultsItemConfidenceMax: () => getTranscriptResponseSentimentAnalysisResultsItemConfidenceMax,
11930
- getTranscriptResponseSentimentAnalysisResultsItemConfidenceMin: () => getTranscriptResponseSentimentAnalysisResultsItemConfidenceMin,
11931
- getTranscriptResponseSpeechModelDefault: () => getTranscriptResponseSpeechModelDefault,
11932
- getTranscriptResponseSpeechThresholdMax: () => getTranscriptResponseSpeechThresholdMax,
11933
- getTranscriptResponseSpeechThresholdMin: () => getTranscriptResponseSpeechThresholdMin,
11934
- getTranscriptResponseUtterancesItemConfidenceMax: () => getTranscriptResponseUtterancesItemConfidenceMax,
11935
- getTranscriptResponseUtterancesItemConfidenceMin: () => getTranscriptResponseUtterancesItemConfidenceMin,
11936
- getTranscriptResponseUtterancesItemWordsItemConfidenceMax: () => getTranscriptResponseUtterancesItemWordsItemConfidenceMax,
11937
- getTranscriptResponseUtterancesItemWordsItemConfidenceMin: () => getTranscriptResponseUtterancesItemWordsItemConfidenceMin,
11938
- getTranscriptResponseWordsItemConfidenceMax: () => getTranscriptResponseWordsItemConfidenceMax,
11939
- getTranscriptResponseWordsItemConfidenceMin: () => getTranscriptResponseWordsItemConfidenceMin,
11836
+ getTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault: () => getTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault,
11837
+ getTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault: () => getTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault,
11838
+ getTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault: () => getTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault,
11839
+ getTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault: () => getTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault,
11840
+ getTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault: () => getTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault,
11841
+ getTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault: () => getTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault,
11940
11842
  getTranscriptSentencesParams: () => getTranscriptSentencesParams,
11941
11843
  getTranscriptSentencesResponse: () => getTranscriptSentencesResponse,
11942
- getTranscriptSentencesResponseConfidenceMax: () => getTranscriptSentencesResponseConfidenceMax,
11943
- getTranscriptSentencesResponseConfidenceMin: () => getTranscriptSentencesResponseConfidenceMin,
11944
- getTranscriptSentencesResponseSentencesItemConfidenceMax: () => getTranscriptSentencesResponseSentencesItemConfidenceMax,
11945
- getTranscriptSentencesResponseSentencesItemConfidenceMin: () => getTranscriptSentencesResponseSentencesItemConfidenceMin,
11946
- getTranscriptSentencesResponseSentencesItemWordsItemConfidenceMax: () => getTranscriptSentencesResponseSentencesItemWordsItemConfidenceMax,
11947
- getTranscriptSentencesResponseSentencesItemWordsItemConfidenceMin: () => getTranscriptSentencesResponseSentencesItemWordsItemConfidenceMin,
11948
- lemurQuestionAnswerBody: () => lemurQuestionAnswerBody,
11949
- lemurQuestionAnswerBodyFinalModelDefault: () => lemurQuestionAnswerBodyFinalModelDefault,
11950
- lemurQuestionAnswerBodyMaxOutputSizeDefault: () => lemurQuestionAnswerBodyMaxOutputSizeDefault,
11951
- lemurQuestionAnswerBodyTemperatureDefault: () => lemurQuestionAnswerBodyTemperatureDefault,
11952
- lemurQuestionAnswerBodyTemperatureMax: () => lemurQuestionAnswerBodyTemperatureMax,
11953
- lemurQuestionAnswerBodyTemperatureMin: () => lemurQuestionAnswerBodyTemperatureMin,
11954
- lemurQuestionAnswerResponse: () => lemurQuestionAnswerResponse,
11955
- lemurQuestionAnswerResponseUsageInputTokensMin: () => lemurQuestionAnswerResponseUsageInputTokensMin,
11956
- lemurQuestionAnswerResponseUsageOutputTokensMin: () => lemurQuestionAnswerResponseUsageOutputTokensMin,
11957
- lemurSummaryBody: () => lemurSummaryBody,
11958
- lemurSummaryBodyFinalModelDefault: () => lemurSummaryBodyFinalModelDefault,
11959
- lemurSummaryBodyMaxOutputSizeDefault: () => lemurSummaryBodyMaxOutputSizeDefault,
11960
- lemurSummaryBodyTemperatureDefault: () => lemurSummaryBodyTemperatureDefault,
11961
- lemurSummaryBodyTemperatureMax: () => lemurSummaryBodyTemperatureMax,
11962
- lemurSummaryBodyTemperatureMin: () => lemurSummaryBodyTemperatureMin,
11963
- lemurSummaryResponse: () => lemurSummaryResponse,
11964
- lemurSummaryResponseUsageInputTokensMin: () => lemurSummaryResponseUsageInputTokensMin,
11965
- lemurSummaryResponseUsageOutputTokensMin: () => lemurSummaryResponseUsageOutputTokensMin,
11966
- lemurTaskBody: () => lemurTaskBody,
11967
- lemurTaskBodyFinalModelDefault: () => lemurTaskBodyFinalModelDefault,
11968
- lemurTaskBodyMaxOutputSizeDefault: () => lemurTaskBodyMaxOutputSizeDefault,
11969
- lemurTaskBodyTemperatureDefault: () => lemurTaskBodyTemperatureDefault,
11970
- lemurTaskBodyTemperatureMax: () => lemurTaskBodyTemperatureMax,
11971
- lemurTaskBodyTemperatureMin: () => lemurTaskBodyTemperatureMin,
11972
- lemurTaskResponse: () => lemurTaskResponse,
11973
- lemurTaskResponseUsageInputTokensMin: () => lemurTaskResponseUsageInputTokensMin,
11974
- lemurTaskResponseUsageOutputTokensMin: () => lemurTaskResponseUsageOutputTokensMin,
11975
11844
  listTranscriptsQueryLimitDefault: () => listTranscriptsQueryLimitDefault,
11976
- listTranscriptsQueryLimitMax: () => listTranscriptsQueryLimitMax,
11977
11845
  listTranscriptsQueryParams: () => listTranscriptsQueryParams,
11978
11846
  listTranscriptsQueryThrottledOnlyDefault: () => listTranscriptsQueryThrottledOnlyDefault,
11979
11847
  listTranscriptsResponse: () => listTranscriptsResponse,
11980
- listTranscriptsResponseTranscriptsItemCompletedRegExp: () => listTranscriptsResponseTranscriptsItemCompletedRegExp,
11981
- listTranscriptsResponseTranscriptsItemCreatedRegExp: () => listTranscriptsResponseTranscriptsItemCreatedRegExp,
11982
- purgeLemurRequestDataParams: () => purgeLemurRequestDataParams,
11983
- purgeLemurRequestDataResponse: () => purgeLemurRequestDataResponse,
11984
11848
  uploadFileResponse: () => uploadFileResponse,
11985
11849
  wordSearchParams: () => wordSearchParams,
11986
11850
  wordSearchQueryParams: () => wordSearchQueryParams,
@@ -11990,43 +11854,192 @@ var import_zod3 = require("zod");
11990
11854
  var uploadFileResponse = import_zod3.z.object({
11991
11855
  upload_url: import_zod3.z.string().describe("A URL that points to your audio file, accessible only by AssemblyAI's servers\n")
11992
11856
  });
11993
- var createTranscriptBodyLanguageCodeDefault = "en_us";
11994
- var createTranscriptBodyLanguageDetectionDefault = false;
11995
- var createTranscriptBodyLanguageConfidenceThresholdDefault = 0;
11996
- var createTranscriptBodyLanguageConfidenceThresholdMin = 0;
11997
- var createTranscriptBodyLanguageConfidenceThresholdMax = 1;
11998
- var createTranscriptBodySpeechModelDefault = "best";
11999
- var createTranscriptBodyPunctuateDefault = true;
12000
- var createTranscriptBodyFormatTextDefault = true;
12001
- var createTranscriptBodyDisfluenciesDefault = false;
12002
- var createTranscriptBodyMultichannelDefault = false;
12003
- var createTranscriptBodyWebhookAuthHeaderNameDefault = null;
12004
- var createTranscriptBodyWebhookAuthHeaderValueDefault = null;
11857
+ var createTranscriptBodyAutoChaptersDefault = false;
12005
11858
  var createTranscriptBodyAutoHighlightsDefault = false;
12006
- var createTranscriptBodyFilterProfanityDefault = false;
12007
- var createTranscriptBodyRedactPiiDefault = false;
12008
- var createTranscriptBodyRedactPiiAudioDefault = false;
12009
- var createTranscriptBodyRedactPiiSubDefault = "hash";
12010
- var createTranscriptBodySpeakerLabelsDefault = false;
12011
- var createTranscriptBodySpeakersExpectedDefault = null;
12012
11859
  var createTranscriptBodyContentSafetyDefault = false;
12013
11860
  var createTranscriptBodyContentSafetyConfidenceDefault = 50;
12014
- var createTranscriptBodyContentSafetyConfidenceMin = 25;
12015
- var createTranscriptBodyContentSafetyConfidenceMax = 100;
11861
+ var createTranscriptBodyDisfluenciesDefault = false;
11862
+ var createTranscriptBodyEntityDetectionDefault = false;
11863
+ var createTranscriptBodyFilterProfanityDefault = false;
11864
+ var createTranscriptBodyFormatTextDefault = true;
12016
11865
  var createTranscriptBodyIabCategoriesDefault = false;
11866
+ var createTranscriptBodyLanguageDetectionDefault = false;
11867
+ var createTranscriptBodyLanguageDetectionOptionsFallbackLanguageDefault = "auto";
11868
+ var createTranscriptBodyLanguageDetectionOptionsCodeSwitchingDefault = false;
11869
+ var createTranscriptBodyLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault = 0.3;
11870
+ var createTranscriptBodyMultichannelDefault = false;
11871
+ var createTranscriptBodyPunctuateDefault = true;
11872
+ var createTranscriptBodyRedactPiiDefault = false;
11873
+ var createTranscriptBodyRedactPiiAudioDefault = false;
11874
+ var createTranscriptBodyRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault = false;
12017
11875
  var createTranscriptBodySentimentAnalysisDefault = false;
12018
- var createTranscriptBodyAutoChaptersDefault = false;
12019
- var createTranscriptBodyEntityDetectionDefault = false;
12020
- var createTranscriptBodySpeechThresholdDefault = 0;
12021
- var createTranscriptBodySpeechThresholdMin = 0;
12022
- var createTranscriptBodySpeechThresholdMax = 1;
11876
+ var createTranscriptBodySpeakerLabelsDefault = false;
11877
+ var createTranscriptBodySpeakerOptionsMinSpeakersExpectedDefault = 1;
11878
+ var createTranscriptBodySpeechUnderstandingRequestTranslationFormalDefault = true;
11879
+ var createTranscriptBodySpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault = false;
12023
11880
  var createTranscriptBodySummarizationDefault = false;
11881
+ var createTranscriptBodyTemperatureDefault = 0;
12024
11882
  var createTranscriptBodyCustomTopicsDefault = false;
12025
11883
  var createTranscriptBody = import_zod3.z.object({
12026
- audio_url: import_zod3.z.string().describe("The URL of the audio or video file to transcribe.")
12027
- }).and(
12028
- import_zod3.z.object({
12029
- language_code: import_zod3.z.enum([
11884
+ audio_end_at: import_zod3.z.number().optional().describe(
11885
+ "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."
11886
+ ),
11887
+ audio_start_from: import_zod3.z.number().optional().describe(
11888
+ "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."
11889
+ ),
11890
+ auto_chapters: import_zod3.z.boolean().optional().describe(
11891
+ "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"
11892
+ ),
11893
+ auto_highlights: import_zod3.z.boolean().optional().describe(
11894
+ "Enable [Key Phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases), either true or false"
11895
+ ),
11896
+ content_safety: import_zod3.z.boolean().optional().describe(
11897
+ "Enable [Content Moderation](https://www.assemblyai.com/docs/content-moderation), can be true or false"
11898
+ ),
11899
+ content_safety_confidence: import_zod3.z.number().default(createTranscriptBodyContentSafetyConfidenceDefault).describe(
11900
+ "The confidence threshold for the [Content Moderation](https://www.assemblyai.com/docs/content-moderation) model. Values must be between 25 and 100."
11901
+ ),
11902
+ custom_spelling: import_zod3.z.array(
11903
+ import_zod3.z.object({
11904
+ from: import_zod3.z.array(import_zod3.z.string()).describe("Words or phrases to replace"),
11905
+ to: import_zod3.z.string().describe("Word to replace with")
11906
+ }).describe(
11907
+ "Object containing words or phrases to replace, and the word or phrase to replace with"
11908
+ )
11909
+ ).optional().describe(
11910
+ "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."
11911
+ ),
11912
+ disfluencies: import_zod3.z.boolean().optional().describe(
11913
+ 'Transcribe [Filler Words](https://www.assemblyai.com/docs/pre-recorded-audio/filler-words), like "umm", in your media file; can be true or false'
11914
+ ),
11915
+ domain: import_zod3.z.string().nullish().describe(
11916
+ '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'
11917
+ ),
11918
+ entity_detection: import_zod3.z.boolean().optional().describe(
11919
+ "Enable [Entity Detection](https://www.assemblyai.com/docs/speech-understanding/entity-detection), can be true or false"
11920
+ ),
11921
+ filter_profanity: import_zod3.z.boolean().optional().describe(
11922
+ "Filter profanity from the transcribed text, can be true or false. See [Profanity Filtering](https://www.assemblyai.com/docs/profanity-filtering) for more details."
11923
+ ),
11924
+ format_text: import_zod3.z.boolean().default(createTranscriptBodyFormatTextDefault).describe(
11925
+ "Enable [Text Formatting](https://www.assemblyai.com/docs/pre-recorded-audio), can be true or false"
11926
+ ),
11927
+ iab_categories: import_zod3.z.boolean().optional().describe(
11928
+ "Enable [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection), can be true or false"
11929
+ ),
11930
+ keyterms_prompt: import_zod3.z.array(import_zod3.z.string()).optional().describe(
11931
+ "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"
11932
+ ),
11933
+ language_code: import_zod3.z.enum([
11934
+ "en",
11935
+ "en_au",
11936
+ "en_uk",
11937
+ "en_us",
11938
+ "es",
11939
+ "fr",
11940
+ "de",
11941
+ "it",
11942
+ "pt",
11943
+ "nl",
11944
+ "af",
11945
+ "sq",
11946
+ "am",
11947
+ "ar",
11948
+ "hy",
11949
+ "as",
11950
+ "az",
11951
+ "ba",
11952
+ "eu",
11953
+ "be",
11954
+ "bn",
11955
+ "bs",
11956
+ "br",
11957
+ "bg",
11958
+ "my",
11959
+ "ca",
11960
+ "zh",
11961
+ "hr",
11962
+ "cs",
11963
+ "da",
11964
+ "et",
11965
+ "fo",
11966
+ "fi",
11967
+ "gl",
11968
+ "ka",
11969
+ "el",
11970
+ "gu",
11971
+ "ht",
11972
+ "ha",
11973
+ "haw",
11974
+ "he",
11975
+ "hi",
11976
+ "hu",
11977
+ "is",
11978
+ "id",
11979
+ "ja",
11980
+ "jw",
11981
+ "kn",
11982
+ "kk",
11983
+ "km",
11984
+ "ko",
11985
+ "lo",
11986
+ "la",
11987
+ "lv",
11988
+ "ln",
11989
+ "lt",
11990
+ "lb",
11991
+ "mk",
11992
+ "mg",
11993
+ "ms",
11994
+ "ml",
11995
+ "mt",
11996
+ "mi",
11997
+ "mr",
11998
+ "mn",
11999
+ "ne",
12000
+ "no",
12001
+ "nn",
12002
+ "oc",
12003
+ "pa",
12004
+ "ps",
12005
+ "fa",
12006
+ "pl",
12007
+ "ro",
12008
+ "ru",
12009
+ "sa",
12010
+ "sr",
12011
+ "sn",
12012
+ "sd",
12013
+ "si",
12014
+ "sk",
12015
+ "sl",
12016
+ "so",
12017
+ "su",
12018
+ "sw",
12019
+ "sv",
12020
+ "tl",
12021
+ "tg",
12022
+ "ta",
12023
+ "tt",
12024
+ "te",
12025
+ "th",
12026
+ "bo",
12027
+ "tr",
12028
+ "tk",
12029
+ "uk",
12030
+ "ur",
12031
+ "uz",
12032
+ "vi",
12033
+ "cy",
12034
+ "yi",
12035
+ "yo"
12036
+ ]).describe(
12037
+ "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"
12038
+ ).or(import_zod3.z.null()).optional().describe(
12039
+ "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"
12040
+ ),
12041
+ language_codes: import_zod3.z.array(
12042
+ import_zod3.z.enum([
12030
12043
  "en",
12031
12044
  "en_au",
12032
12045
  "en_uk",
@@ -12130,53 +12143,360 @@ var createTranscriptBody = import_zod3.z.object({
12130
12143
  "yi",
12131
12144
  "yo"
12132
12145
  ]).describe(
12133
- "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"
12134
- ).or(import_zod3.z.string()).or(import_zod3.z.null()).default(createTranscriptBodyLanguageCodeDefault).describe(
12135
- "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"
12136
- ),
12137
- language_detection: import_zod3.z.boolean().optional().describe(
12138
- "Enable [Automatic language detection](https://www.assemblyai.com/docs/models/speech-recognition#automatic-language-detection), either true or false."
12139
- ),
12140
- language_confidence_threshold: import_zod3.z.number().min(createTranscriptBodyLanguageConfidenceThresholdMin).max(createTranscriptBodyLanguageConfidenceThresholdMax).optional().describe(
12141
- "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"
12142
- ),
12143
- 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(
12144
- 'The speech model to use for the transcription. When `null`, the "best" model is used.'
12145
- ),
12146
- punctuate: import_zod3.z.boolean().default(createTranscriptBodyPunctuateDefault).describe("Enable Automatic Punctuation, can be true or false"),
12147
- format_text: import_zod3.z.boolean().default(createTranscriptBodyFormatTextDefault).describe("Enable Text Formatting, can be true or false"),
12148
- disfluencies: import_zod3.z.boolean().optional().describe(
12149
- 'Transcribe Filler Words, like "umm", in your media file; can be true or false'
12150
- ),
12151
- multichannel: import_zod3.z.boolean().optional().describe(
12152
- "Enable [Multichannel](https://www.assemblyai.com/docs/models/speech-recognition#multichannel-transcription) transcription, can be true or false."
12153
- ),
12154
- webhook_url: import_zod3.z.string().optional().describe(
12155
- "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"
12146
+ "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"
12147
+ )
12148
+ ).nullish().describe(
12149
+ "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"
12150
+ ),
12151
+ language_confidence_threshold: import_zod3.z.number().optional().describe(
12152
+ "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"
12153
+ ),
12154
+ language_detection: import_zod3.z.boolean().optional().describe(
12155
+ "Enable [Automatic language detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection), either true or false."
12156
+ ),
12157
+ language_detection_options: import_zod3.z.object({
12158
+ expected_languages: import_zod3.z.array(import_zod3.z.string()).optional().describe(
12159
+ '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.'
12156
12160
  ),
12157
- webhook_auth_header_name: import_zod3.z.string().nullish().describe(
12158
- "The header name to be sent with the transcript completed or failed webhook requests"
12161
+ fallback_language: import_zod3.z.string().default(createTranscriptBodyLanguageDetectionOptionsFallbackLanguageDefault).describe(
12162
+ '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'
12159
12163
  ),
12160
- webhook_auth_header_value: import_zod3.z.string().nullish().describe(
12161
- "The header value to send back with the transcript completed or failed webhook requests for added security"
12164
+ code_switching: import_zod3.z.boolean().optional().describe(
12165
+ "Whether [code switching](/docs/speech-to-text/pre-recorded-audio/code-switching) should be detected.\n"
12162
12166
  ),
12163
- auto_highlights: import_zod3.z.boolean().optional().describe("Enable Key Phrases, either true or false"),
12164
- audio_start_from: import_zod3.z.number().optional().describe("The point in time, in milliseconds, to begin transcribing in your media file"),
12165
- audio_end_at: import_zod3.z.number().optional().describe("The point in time, in milliseconds, to stop transcribing in your media file"),
12166
- word_boost: import_zod3.z.array(import_zod3.z.string()).optional().describe("The list of custom vocabulary to boost transcription probability for"),
12167
- boost_param: import_zod3.z.enum(["low", "default", "high"]).optional().describe("How much to boost specified words"),
12168
- filter_profanity: import_zod3.z.boolean().optional().describe("Filter profanity from the transcribed text, can be true or false"),
12169
- redact_pii: import_zod3.z.boolean().optional().describe(
12170
- "Redact PII from the transcribed text using the Redact PII model, can be true or false"
12167
+ code_switching_confidence_threshold: import_zod3.z.number().default(
12168
+ createTranscriptBodyLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault
12169
+ ).describe(
12170
+ "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"
12171
+ )
12172
+ }).optional().describe(
12173
+ "Specify options for [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection)."
12174
+ ),
12175
+ multichannel: import_zod3.z.boolean().optional().describe(
12176
+ "Enable [Multichannel](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) transcription, can be true or false."
12177
+ ),
12178
+ prompt: import_zod3.z.string().optional().describe(
12179
+ "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"
12180
+ ),
12181
+ punctuate: import_zod3.z.boolean().default(createTranscriptBodyPunctuateDefault).describe(
12182
+ "Enable [Automatic Punctuation](https://www.assemblyai.com/docs/pre-recorded-audio), can be true or false"
12183
+ ),
12184
+ redact_pii: import_zod3.z.boolean().optional().describe(
12185
+ "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."
12186
+ ),
12187
+ redact_pii_audio: import_zod3.z.boolean().optional().describe(
12188
+ '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.'
12189
+ ),
12190
+ redact_pii_audio_options: import_zod3.z.object({
12191
+ return_redacted_no_speech_audio: import_zod3.z.boolean().optional().describe(
12192
+ "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`."
12171
12193
  ),
12172
- redact_pii_audio: import_zod3.z.boolean().optional().describe(
12173
- '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.'
12194
+ override_audio_redaction_method: import_zod3.z.enum(["silence"]).optional().describe(
12195
+ "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."
12196
+ )
12197
+ }).optional().describe(
12198
+ "Specify options for [PII redacted audio](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) files."
12199
+ ),
12200
+ redact_pii_audio_quality: import_zod3.z.enum(["mp3", "wav"]).optional().describe(
12201
+ "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."
12202
+ ),
12203
+ redact_pii_policies: import_zod3.z.array(
12204
+ import_zod3.z.enum([
12205
+ "account_number",
12206
+ "banking_information",
12207
+ "blood_type",
12208
+ "credit_card_cvv",
12209
+ "credit_card_expiration",
12210
+ "credit_card_number",
12211
+ "date",
12212
+ "date_interval",
12213
+ "date_of_birth",
12214
+ "drivers_license",
12215
+ "drug",
12216
+ "duration",
12217
+ "email_address",
12218
+ "event",
12219
+ "filename",
12220
+ "gender_sexuality",
12221
+ "healthcare_number",
12222
+ "injury",
12223
+ "ip_address",
12224
+ "language",
12225
+ "location",
12226
+ "marital_status",
12227
+ "medical_condition",
12228
+ "medical_process",
12229
+ "money_amount",
12230
+ "nationality",
12231
+ "number_sequence",
12232
+ "occupation",
12233
+ "organization",
12234
+ "passport_number",
12235
+ "password",
12236
+ "person_age",
12237
+ "person_name",
12238
+ "phone_number",
12239
+ "physical_attribute",
12240
+ "political_affiliation",
12241
+ "religion",
12242
+ "statistics",
12243
+ "time",
12244
+ "url",
12245
+ "us_social_security_number",
12246
+ "username",
12247
+ "vehicle_id",
12248
+ "zodiac_sign"
12249
+ ]).describe("The type of PII to redact")
12250
+ ).optional().describe(
12251
+ "The list of PII Redaction policies to enable. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction) for more details."
12252
+ ),
12253
+ redact_pii_sub: import_zod3.z.enum(["entity_name", "hash"]).describe(
12254
+ "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."
12255
+ ).or(import_zod3.z.null()).optional().describe(
12256
+ "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."
12257
+ ),
12258
+ sentiment_analysis: import_zod3.z.boolean().optional().describe(
12259
+ "Enable [Sentiment Analysis](https://www.assemblyai.com/docs/speech-understanding/sentiment-analysis), can be true or false"
12260
+ ),
12261
+ speaker_labels: import_zod3.z.boolean().optional().describe(
12262
+ "Enable [Speaker diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization), can be true or false"
12263
+ ),
12264
+ speaker_options: import_zod3.z.object({
12265
+ min_speakers_expected: import_zod3.z.number().default(createTranscriptBodySpeakerOptionsMinSpeakersExpectedDefault).describe(
12266
+ "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."
12174
12267
  ),
12175
- redact_pii_audio_quality: import_zod3.z.enum(["mp3", "wav"]).optional().describe(
12176
- "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."
12268
+ max_speakers_expected: import_zod3.z.number().optional().describe(
12269
+ "<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"
12270
+ )
12271
+ }).optional().describe(
12272
+ "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."
12273
+ ),
12274
+ speakers_expected: import_zod3.z.number().nullish().describe(
12275
+ "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."
12276
+ ),
12277
+ speech_models: import_zod3.z.array(
12278
+ import_zod3.z.string().describe(
12279
+ "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."
12280
+ )
12281
+ ).describe(
12282
+ "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"
12283
+ ),
12284
+ speech_threshold: import_zod3.z.number().nullish().describe(
12285
+ "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"
12286
+ ),
12287
+ speech_understanding: import_zod3.z.object({
12288
+ request: import_zod3.z.object({
12289
+ translation: import_zod3.z.object({
12290
+ target_languages: import_zod3.z.array(import_zod3.z.string()).describe(
12291
+ 'List of target language codes (e.g., `["es", "de"]`). See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for supported languages.'
12292
+ ),
12293
+ formal: import_zod3.z.boolean().default(createTranscriptBodySpeechUnderstandingRequestTranslationFormalDefault).describe(
12294
+ "Use formal language style. See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for more details."
12295
+ ),
12296
+ match_original_utterance: import_zod3.z.boolean().optional().describe(
12297
+ "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."
12298
+ )
12299
+ })
12300
+ }).describe(
12301
+ "Request body for [Translation](https://www.assemblyai.com/docs/speech-understanding/translation)."
12302
+ ).or(
12303
+ import_zod3.z.object({
12304
+ speaker_identification: import_zod3.z.object({
12305
+ speaker_type: import_zod3.z.enum(["role", "name"]).describe(
12306
+ "Type of speaker identification. See [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification) for details on each type."
12307
+ ),
12308
+ known_values: import_zod3.z.array(import_zod3.z.string()).optional().describe(
12309
+ 'Required if speaker_type is "role". Each value must be 35 characters or less.'
12310
+ ),
12311
+ speakers: import_zod3.z.array(
12312
+ import_zod3.z.object({
12313
+ role: import_zod3.z.string().optional().describe(
12314
+ 'The role of the speaker. Required when `speaker_type` is "role".'
12315
+ ),
12316
+ name: import_zod3.z.string().optional().describe(
12317
+ 'The name of the speaker. Required when `speaker_type` is "name".'
12318
+ ),
12319
+ description: import_zod3.z.string().optional().describe(
12320
+ "A description of the speaker to help the model identify them based on conversational context."
12321
+ )
12322
+ })
12323
+ ).optional().describe(
12324
+ "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."
12325
+ )
12326
+ })
12327
+ }).describe(
12328
+ "Request body for [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification)."
12329
+ )
12330
+ ).or(
12331
+ import_zod3.z.object({
12332
+ custom_formatting: import_zod3.z.object({
12333
+ date: import_zod3.z.string().optional().describe(
12334
+ 'Date format pattern (e.g., `"mm/dd/yyyy"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
12335
+ ),
12336
+ phone_number: import_zod3.z.string().optional().describe(
12337
+ 'Phone number format pattern (e.g., `"(xxx)xxx-xxxx"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
12338
+ ),
12339
+ email: import_zod3.z.string().optional().describe(
12340
+ 'Email format pattern (e.g., `"username@domain.com"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
12341
+ )
12342
+ })
12343
+ }).describe(
12344
+ "Request body for [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting)."
12345
+ )
12346
+ )
12347
+ }).optional().describe(
12348
+ "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"
12349
+ ),
12350
+ summarization: import_zod3.z.boolean().optional().describe(
12351
+ "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"
12352
+ ),
12353
+ summary_model: import_zod3.z.enum(["informative", "conversational", "catchy"]).optional().describe("The model to summarize the transcript"),
12354
+ summary_type: import_zod3.z.enum(["bullets", "bullets_verbose", "gist", "headline", "paragraph"]).optional().describe("The type of summary"),
12355
+ remove_audio_tags: import_zod3.z.enum(["all"]).describe(
12356
+ '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'
12357
+ ).or(import_zod3.z.null()).optional().describe(
12358
+ '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'
12359
+ ),
12360
+ temperature: import_zod3.z.number().optional().describe(
12361
+ "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"
12362
+ ),
12363
+ webhook_auth_header_name: import_zod3.z.string().nullish().describe(
12364
+ "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"
12365
+ ),
12366
+ webhook_auth_header_value: import_zod3.z.string().nullish().describe(
12367
+ "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"
12368
+ ),
12369
+ webhook_url: import_zod3.z.string().optional().describe(
12370
+ "The URL to which we send [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) requests.\n"
12371
+ ),
12372
+ custom_topics: import_zod3.z.boolean().optional().describe("This parameter does not currently have any functionality attached to it."),
12373
+ topics: import_zod3.z.array(import_zod3.z.string()).optional().describe("This parameter does not currently have any functionality attached to it."),
12374
+ audio_url: import_zod3.z.string().describe("The URL of the audio or video file to transcribe.")
12375
+ }).describe("The parameters for creating a transcript");
12376
+ var createTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault = "auto";
12377
+ var createTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault = false;
12378
+ var createTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault = 0.3;
12379
+ var createTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault = false;
12380
+ var createTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault = true;
12381
+ var createTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault = false;
12382
+ var createTranscriptResponse = import_zod3.z.object({
12383
+ audio_channels: import_zod3.z.number().optional().describe(
12384
+ "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."
12385
+ ),
12386
+ audio_duration: import_zod3.z.number().nullish().describe("The duration of this transcript object's media file, in seconds"),
12387
+ audio_end_at: import_zod3.z.number().nullish().describe(
12388
+ "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."
12389
+ ),
12390
+ audio_start_from: import_zod3.z.number().nullish().describe(
12391
+ "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."
12392
+ ),
12393
+ audio_url: import_zod3.z.string().describe("The URL of the media that was transcribed"),
12394
+ auto_chapters: import_zod3.z.boolean().nullish().describe(
12395
+ "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"
12396
+ ),
12397
+ auto_highlights: import_zod3.z.boolean().describe(
12398
+ "Whether [Key Phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases) is enabled, either true or false"
12399
+ ),
12400
+ auto_highlights_result: import_zod3.z.object({
12401
+ status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
12402
+ results: import_zod3.z.array(
12403
+ import_zod3.z.object({
12404
+ count: import_zod3.z.number().describe("The total number of times the key phrase appears in the audio file"),
12405
+ rank: import_zod3.z.number().describe(
12406
+ "The total relevancy to the overall audio file of this key phrase - a greater number means more relevant"
12407
+ ),
12408
+ text: import_zod3.z.string().describe("The text itself of the key phrase"),
12409
+ timestamps: import_zod3.z.array(
12410
+ import_zod3.z.object({
12411
+ start: import_zod3.z.number().describe("The start time in milliseconds"),
12412
+ end: import_zod3.z.number().describe("The end time in milliseconds")
12413
+ }).describe("Timestamp containing a start and end property in milliseconds")
12414
+ ).describe("The timestamp of the of the key phrase")
12415
+ })
12416
+ ).describe("A temporally-sequential array of Key Phrases")
12417
+ }).describe(
12418
+ "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"
12419
+ ).or(import_zod3.z.null()).optional().describe(
12420
+ "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"
12421
+ ),
12422
+ chapters: import_zod3.z.array(
12423
+ import_zod3.z.object({
12424
+ gist: import_zod3.z.string().describe(
12425
+ "An ultra-short summary (just a few words) of the content spoken in the chapter"
12426
+ ),
12427
+ headline: import_zod3.z.string().describe("A single sentence summary of the content spoken during the chapter"),
12428
+ summary: import_zod3.z.string().describe("A one paragraph summary of the content spoken during the chapter"),
12429
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter"),
12430
+ end: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter")
12431
+ }).describe("Chapter of the audio file")
12432
+ ).nullish().describe(
12433
+ "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."
12434
+ ),
12435
+ confidence: import_zod3.z.number().nullish().describe(
12436
+ "The confidence score for the transcript, between 0.0 (low confidence) and 1.0 (high confidence)"
12437
+ ),
12438
+ content_safety: import_zod3.z.boolean().nullish().describe(
12439
+ "Whether [Content Moderation](https://www.assemblyai.com/docs/content-moderation) is enabled, can be true or false"
12440
+ ),
12441
+ content_safety_labels: import_zod3.z.object({
12442
+ status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
12443
+ results: import_zod3.z.array(
12444
+ import_zod3.z.object({
12445
+ text: import_zod3.z.string().describe("The transcript of the section flagged by the Content Moderation model"),
12446
+ labels: import_zod3.z.array(
12447
+ import_zod3.z.object({
12448
+ label: import_zod3.z.string().describe("The label of the sensitive topic"),
12449
+ confidence: import_zod3.z.number().describe("The confidence score for the topic being discussed, from 0 to 1"),
12450
+ severity: import_zod3.z.number().describe("How severely the topic is discussed in the section, from 0 to 1")
12451
+ })
12452
+ ).describe(
12453
+ "An array of safety labels, one per sensitive topic that was detected in the section"
12454
+ ),
12455
+ sentences_idx_start: import_zod3.z.number().describe("The sentence index at which the section begins"),
12456
+ sentences_idx_end: import_zod3.z.number().describe("The sentence index at which the section ends"),
12457
+ timestamp: import_zod3.z.object({
12458
+ start: import_zod3.z.number().describe("The start time in milliseconds"),
12459
+ end: import_zod3.z.number().describe("The end time in milliseconds")
12460
+ }).describe("Timestamp containing a start and end property in milliseconds")
12461
+ })
12462
+ ).describe("An array of results for the Content Moderation model"),
12463
+ summary: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.number()).describe(
12464
+ "A summary of the Content Moderation confidence results for the entire audio file"
12177
12465
  ),
12178
- redact_pii_policies: import_zod3.z.array(
12179
- import_zod3.z.enum([
12466
+ severity_score_summary: import_zod3.z.record(
12467
+ import_zod3.z.string(),
12468
+ import_zod3.z.object({
12469
+ low: import_zod3.z.number(),
12470
+ medium: import_zod3.z.number(),
12471
+ high: import_zod3.z.number()
12472
+ })
12473
+ ).describe(
12474
+ "A summary of the Content Moderation severity results for the entire audio file"
12475
+ )
12476
+ }).describe(
12477
+ "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"
12478
+ ).or(import_zod3.z.null()).optional().describe(
12479
+ "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"
12480
+ ),
12481
+ custom_spelling: import_zod3.z.array(
12482
+ import_zod3.z.object({
12483
+ from: import_zod3.z.array(import_zod3.z.string()).describe("Words or phrases to replace"),
12484
+ to: import_zod3.z.string().describe("Word to replace with")
12485
+ }).describe(
12486
+ "Object containing words or phrases to replace, and the word or phrase to replace with"
12487
+ )
12488
+ ).nullish().describe(
12489
+ "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."
12490
+ ),
12491
+ disfluencies: import_zod3.z.boolean().nullish().describe(
12492
+ 'Transcribe [Filler Words](https://www.assemblyai.com/docs/pre-recorded-audio/filler-words), like "umm", in your media file; can be true or false'
12493
+ ),
12494
+ domain: import_zod3.z.string().nullish().describe(
12495
+ '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'
12496
+ ),
12497
+ entities: import_zod3.z.array(
12498
+ import_zod3.z.object({
12499
+ entity_type: import_zod3.z.enum([
12180
12500
  "account_number",
12181
12501
  "banking_information",
12182
12502
  "blood_type",
@@ -12221,103 +12541,59 @@ var createTranscriptBody = import_zod3.z.object({
12221
12541
  "username",
12222
12542
  "vehicle_id",
12223
12543
  "zodiac_sign"
12224
- ]).describe("The type of PII to redact")
12225
- ).optional().describe(
12226
- "The list of PII Redaction policies to enable. See [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more details."
12227
- ),
12228
- redact_pii_sub: import_zod3.z.enum(["entity_name", "hash"]).describe(
12229
- '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.'
12230
- ).or(import_zod3.z.null()).default(createTranscriptBodyRedactPiiSubDefault).describe(
12231
- '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.'
12232
- ),
12233
- speaker_labels: import_zod3.z.boolean().optional().describe(
12234
- "Enable [Speaker diarization](https://www.assemblyai.com/docs/models/speaker-diarization), can be true or false"
12235
- ),
12236
- speakers_expected: import_zod3.z.number().nullish().describe(
12237
- "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."
12238
- ),
12239
- content_safety: import_zod3.z.boolean().optional().describe(
12240
- "Enable [Content Moderation](https://www.assemblyai.com/docs/models/content-moderation), can be true or false"
12241
- ),
12242
- content_safety_confidence: import_zod3.z.number().min(createTranscriptBodyContentSafetyConfidenceMin).max(createTranscriptBodyContentSafetyConfidenceMax).default(createTranscriptBodyContentSafetyConfidenceDefault).describe(
12243
- "The confidence threshold for the Content Moderation model. Values must be between 25 and 100."
12244
- ),
12245
- iab_categories: import_zod3.z.boolean().optional().describe(
12246
- "Enable [Topic Detection](https://www.assemblyai.com/docs/models/topic-detection), can be true or false"
12247
- ),
12248
- custom_spelling: import_zod3.z.array(
12249
- import_zod3.z.object({
12250
- from: import_zod3.z.array(import_zod3.z.string().describe("Word or phrase to replace")).describe("Words or phrases to replace"),
12251
- to: import_zod3.z.string().describe("Word to replace with")
12252
- }).describe(
12253
- "Object containing words or phrases to replace, and the word or phrase to replace with"
12544
+ ]).describe("The type of entity for the detected entity"),
12545
+ text: import_zod3.z.string().describe("The text for the detected entity"),
12546
+ start: import_zod3.z.number().describe(
12547
+ "The starting time, in milliseconds, at which the detected entity appears in the audio file"
12548
+ ),
12549
+ end: import_zod3.z.number().describe(
12550
+ "The ending time, in milliseconds, for the detected entity in the audio file"
12254
12551
  )
12255
- ).optional().describe("Customize how words are spelled and formatted using to and from values"),
12256
- keyterms_prompt: import_zod3.z.array(import_zod3.z.string()).optional().describe(
12257
- "<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"
12258
- ),
12259
- prompt: import_zod3.z.string().optional().describe("This parameter does not currently have any functionality attached to it."),
12260
- sentiment_analysis: import_zod3.z.boolean().optional().describe(
12261
- "Enable [Sentiment Analysis](https://www.assemblyai.com/docs/models/sentiment-analysis), can be true or false"
12262
- ),
12263
- auto_chapters: import_zod3.z.boolean().optional().describe(
12264
- "Enable [Auto Chapters](https://www.assemblyai.com/docs/models/auto-chapters), can be true or false"
12265
- ),
12266
- entity_detection: import_zod3.z.boolean().optional().describe(
12267
- "Enable [Entity Detection](https://www.assemblyai.com/docs/models/entity-detection), can be true or false"
12268
- ),
12269
- speech_threshold: import_zod3.z.number().min(createTranscriptBodySpeechThresholdMin).max(createTranscriptBodySpeechThresholdMax).nullish().describe(
12270
- "Reject audio files that contain less than this fraction of speech.\nValid values are in the range [0, 1] inclusive.\n"
12271
- ),
12272
- summarization: import_zod3.z.boolean().optional().describe(
12273
- "Enable [Summarization](https://www.assemblyai.com/docs/models/summarization), can be true or false"
12274
- ),
12275
- summary_model: import_zod3.z.enum(["informative", "conversational", "catchy"]).optional().describe("The model to summarize the transcript"),
12276
- summary_type: import_zod3.z.enum(["bullets", "bullets_verbose", "gist", "headline", "paragraph"]).optional().describe("The type of summary"),
12277
- custom_topics: import_zod3.z.boolean().optional().describe("Enable custom topics, either true or false"),
12278
- topics: import_zod3.z.array(import_zod3.z.string()).optional().describe("The list of custom topics")
12279
- }).describe("The parameters for creating a transcript")
12280
- ).describe("The parameters for creating a transcript");
12281
- var createTranscriptResponseLanguageConfidenceThresholdMin = 0;
12282
- var createTranscriptResponseLanguageConfidenceThresholdMax = 1;
12283
- var createTranscriptResponseLanguageConfidenceMin = 0;
12284
- var createTranscriptResponseLanguageConfidenceMax = 1;
12285
- var createTranscriptResponseSpeechModelDefault = null;
12286
- var createTranscriptResponseWordsItemConfidenceMin = 0;
12287
- var createTranscriptResponseWordsItemConfidenceMax = 1;
12288
- var createTranscriptResponseUtterancesItemConfidenceMin = 0;
12289
- var createTranscriptResponseUtterancesItemConfidenceMax = 1;
12290
- var createTranscriptResponseUtterancesItemWordsItemConfidenceMin = 0;
12291
- var createTranscriptResponseUtterancesItemWordsItemConfidenceMax = 1;
12292
- var createTranscriptResponseConfidenceMin = 0;
12293
- var createTranscriptResponseConfidenceMax = 1;
12294
- var createTranscriptResponseAutoHighlightsResultResultsItemRankMin = 0;
12295
- var createTranscriptResponseAutoHighlightsResultResultsItemRankMax = 1;
12296
- var createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin = 0;
12297
- var createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax = 1;
12298
- var createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin = 0;
12299
- var createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax = 1;
12300
- var createTranscriptResponseContentSafetyLabelsSummaryMinOne = 0;
12301
- var createTranscriptResponseContentSafetyLabelsSummaryMaxOne = 1;
12302
- var createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin = 0;
12303
- var createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax = 1;
12304
- var createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin = 0;
12305
- var createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax = 1;
12306
- var createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin = 0;
12307
- var createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax = 1;
12308
- var createTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin = 0;
12309
- var createTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax = 1;
12310
- var createTranscriptResponseIabCategoriesResultSummaryMinOne = 0;
12311
- var createTranscriptResponseIabCategoriesResultSummaryMaxOne = 1;
12312
- var createTranscriptResponseSentimentAnalysisResultsItemConfidenceMin = 0;
12313
- var createTranscriptResponseSentimentAnalysisResultsItemConfidenceMax = 1;
12314
- var createTranscriptResponseSpeechThresholdMin = 0;
12315
- var createTranscriptResponseSpeechThresholdMax = 1;
12316
- var createTranscriptResponse = import_zod3.z.object({
12552
+ }).describe("A detected entity")
12553
+ ).nullish().describe(
12554
+ "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"
12555
+ ),
12556
+ entity_detection: import_zod3.z.boolean().nullish().describe(
12557
+ "Whether [Entity Detection](https://www.assemblyai.com/docs/speech-understanding/entity-detection) is enabled, can be true or false"
12558
+ ),
12559
+ error: import_zod3.z.string().optional().describe("Error message of why the transcript failed"),
12560
+ filter_profanity: import_zod3.z.boolean().nullish().describe(
12561
+ "Whether [Profanity Filtering](https://www.assemblyai.com/docs/profanity-filtering) is enabled, either true or false"
12562
+ ),
12563
+ format_text: import_zod3.z.boolean().nullish().describe(
12564
+ "Whether [Text Formatting](https://www.assemblyai.com/docs/pre-recorded-audio) is enabled, either true or false"
12565
+ ),
12566
+ iab_categories: import_zod3.z.boolean().nullish().describe(
12567
+ "Whether [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection) is enabled, can be true or false"
12568
+ ),
12569
+ iab_categories_result: import_zod3.z.object({
12570
+ status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
12571
+ results: import_zod3.z.array(
12572
+ import_zod3.z.object({
12573
+ text: import_zod3.z.string().describe("The text in the transcript in which a detected topic occurs"),
12574
+ labels: import_zod3.z.array(
12575
+ import_zod3.z.object({
12576
+ relevance: import_zod3.z.number().describe("How relevant the detected topic is of a detected topic"),
12577
+ label: import_zod3.z.string().describe(
12578
+ "The IAB taxonomical label for the label of the detected topic, where > denotes supertopic/subtopic relationship"
12579
+ )
12580
+ })
12581
+ ).optional().describe("An array of detected topics in the text"),
12582
+ timestamp: import_zod3.z.object({
12583
+ start: import_zod3.z.number().describe("The start time in milliseconds"),
12584
+ end: import_zod3.z.number().describe("The end time in milliseconds")
12585
+ }).optional().describe("Timestamp containing a start and end property in milliseconds")
12586
+ }).describe("The result of the topic detection model")
12587
+ ).describe("An array of results for the Topic Detection model"),
12588
+ summary: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.number()).describe("The overall relevance of topic to the entire audio file")
12589
+ }).describe(
12590
+ "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"
12591
+ ).or(import_zod3.z.null()).optional().describe(
12592
+ "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"
12593
+ ),
12317
12594
  id: import_zod3.z.string().uuid().describe("The unique identifier of your transcript"),
12318
- audio_url: import_zod3.z.string().describe("The URL of the media that was transcribed"),
12319
- status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).describe(
12320
- "The status of your transcript. Possible values are queued, processing, completed, or error."
12595
+ keyterms_prompt: import_zod3.z.array(import_zod3.z.string()).optional().describe(
12596
+ "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"
12321
12597
  ),
12322
12598
  language_code: import_zod3.z.enum([
12323
12599
  "en",
@@ -12422,138 +12698,175 @@ var createTranscriptResponse = import_zod3.z.object({
12422
12698
  "cy",
12423
12699
  "yi",
12424
12700
  "yo"
12425
- ]).describe(
12426
- "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"
12427
- ).or(import_zod3.z.string()).optional().describe(
12428
- "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"
12429
- ),
12430
- language_detection: import_zod3.z.boolean().nullish().describe(
12431
- "Whether [Automatic language detection](https://www.assemblyai.com/docs/models/speech-recognition#automatic-language-detection) is enabled, either true or false"
12432
- ),
12433
- language_confidence_threshold: import_zod3.z.number().min(createTranscriptResponseLanguageConfidenceThresholdMin).max(createTranscriptResponseLanguageConfidenceThresholdMax).nullable().describe(
12434
- "The confidence threshold for the automatically detected language.\nAn error will be returned if the language confidence is below this threshold.\n"
12435
- ),
12436
- language_confidence: import_zod3.z.number().min(createTranscriptResponseLanguageConfidenceMin).max(createTranscriptResponseLanguageConfidenceMax).nullable().describe(
12437
- "The confidence score for the detected language, between 0.0 (low confidence) and 1.0 (high confidence)"
12438
- ),
12439
- 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(
12440
- "The speech model used for the transcription. When `null`, the default model is used."
12441
- ),
12442
- text: import_zod3.z.string().nullish().describe("The textual transcript of your media file"),
12443
- words: import_zod3.z.array(
12444
- import_zod3.z.object({
12445
- confidence: import_zod3.z.number().min(createTranscriptResponseWordsItemConfidenceMin).max(createTranscriptResponseWordsItemConfidenceMax).describe("The confidence score for the transcript of this word"),
12446
- start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
12447
- end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
12448
- text: import_zod3.z.string().describe("The text of the word"),
12449
- channel: import_zod3.z.string().nullish().describe(
12450
- "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
12451
- ),
12452
- speaker: import_zod3.z.string().nullable().describe(
12453
- "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
12454
- )
12455
- })
12456
- ).nullish().describe(
12457
- "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"
12701
+ ]).optional().describe(
12702
+ "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"
12458
12703
  ),
12459
- utterances: import_zod3.z.array(
12460
- import_zod3.z.object({
12461
- confidence: import_zod3.z.number().min(createTranscriptResponseUtterancesItemConfidenceMin).max(createTranscriptResponseUtterancesItemConfidenceMax).describe("The confidence score for the transcript of this utterance"),
12462
- start: import_zod3.z.number().describe("The starting time, in milliseconds, of the utterance in the audio file"),
12463
- end: import_zod3.z.number().describe("The ending time, in milliseconds, of the utterance in the audio file"),
12464
- text: import_zod3.z.string().describe("The text for this utterance"),
12465
- words: import_zod3.z.array(
12466
- import_zod3.z.object({
12467
- confidence: import_zod3.z.number().min(createTranscriptResponseUtterancesItemWordsItemConfidenceMin).max(createTranscriptResponseUtterancesItemWordsItemConfidenceMax).describe("The confidence score for the transcript of this word"),
12468
- start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
12469
- end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
12470
- text: import_zod3.z.string().describe("The text of the word"),
12471
- channel: import_zod3.z.string().nullish().describe(
12472
- "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
12473
- ),
12474
- speaker: import_zod3.z.string().nullable().describe(
12475
- "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
12476
- )
12477
- })
12478
- ).describe("The words in the utterance."),
12479
- channel: import_zod3.z.string().nullish().describe(
12480
- "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
12481
- ),
12482
- speaker: import_zod3.z.string().describe(
12483
- '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.'
12484
- )
12485
- })
12704
+ language_codes: import_zod3.z.array(
12705
+ import_zod3.z.enum([
12706
+ "en",
12707
+ "en_au",
12708
+ "en_uk",
12709
+ "en_us",
12710
+ "es",
12711
+ "fr",
12712
+ "de",
12713
+ "it",
12714
+ "pt",
12715
+ "nl",
12716
+ "af",
12717
+ "sq",
12718
+ "am",
12719
+ "ar",
12720
+ "hy",
12721
+ "as",
12722
+ "az",
12723
+ "ba",
12724
+ "eu",
12725
+ "be",
12726
+ "bn",
12727
+ "bs",
12728
+ "br",
12729
+ "bg",
12730
+ "my",
12731
+ "ca",
12732
+ "zh",
12733
+ "hr",
12734
+ "cs",
12735
+ "da",
12736
+ "et",
12737
+ "fo",
12738
+ "fi",
12739
+ "gl",
12740
+ "ka",
12741
+ "el",
12742
+ "gu",
12743
+ "ht",
12744
+ "ha",
12745
+ "haw",
12746
+ "he",
12747
+ "hi",
12748
+ "hu",
12749
+ "is",
12750
+ "id",
12751
+ "ja",
12752
+ "jw",
12753
+ "kn",
12754
+ "kk",
12755
+ "km",
12756
+ "ko",
12757
+ "lo",
12758
+ "la",
12759
+ "lv",
12760
+ "ln",
12761
+ "lt",
12762
+ "lb",
12763
+ "mk",
12764
+ "mg",
12765
+ "ms",
12766
+ "ml",
12767
+ "mt",
12768
+ "mi",
12769
+ "mr",
12770
+ "mn",
12771
+ "ne",
12772
+ "no",
12773
+ "nn",
12774
+ "oc",
12775
+ "pa",
12776
+ "ps",
12777
+ "fa",
12778
+ "pl",
12779
+ "ro",
12780
+ "ru",
12781
+ "sa",
12782
+ "sr",
12783
+ "sn",
12784
+ "sd",
12785
+ "si",
12786
+ "sk",
12787
+ "sl",
12788
+ "so",
12789
+ "su",
12790
+ "sw",
12791
+ "sv",
12792
+ "tl",
12793
+ "tg",
12794
+ "ta",
12795
+ "tt",
12796
+ "te",
12797
+ "th",
12798
+ "bo",
12799
+ "tr",
12800
+ "tk",
12801
+ "uk",
12802
+ "ur",
12803
+ "uz",
12804
+ "vi",
12805
+ "cy",
12806
+ "yi",
12807
+ "yo"
12808
+ ]).describe(
12809
+ "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"
12810
+ )
12486
12811
  ).nullish().describe(
12487
- "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"
12812
+ "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"
12488
12813
  ),
12489
- confidence: import_zod3.z.number().min(createTranscriptResponseConfidenceMin).max(createTranscriptResponseConfidenceMax).nullish().describe(
12490
- "The confidence score for the transcript, between 0.0 (low confidence) and 1.0 (high confidence)"
12814
+ language_confidence: import_zod3.z.number().nullable().describe(
12815
+ "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."
12491
12816
  ),
12492
- audio_duration: import_zod3.z.number().nullish().describe("The duration of this transcript object's media file, in seconds"),
12493
- punctuate: import_zod3.z.boolean().nullish().describe("Whether Automatic Punctuation is enabled, either true or false"),
12494
- format_text: import_zod3.z.boolean().nullish().describe("Whether Text Formatting is enabled, either true or false"),
12495
- disfluencies: import_zod3.z.boolean().nullish().describe('Transcribe Filler Words, like "umm", in your media file; can be true or false'),
12496
- multichannel: import_zod3.z.boolean().nullish().describe(
12497
- "Whether [Multichannel transcription](https://www.assemblyai.com/docs/models/speech-recognition#multichannel-transcription) was enabled in the transcription request, either true or false"
12498
- ),
12499
- audio_channels: import_zod3.z.number().optional().describe(
12500
- "The number of audio channels in the audio file. This is only present when multichannel is enabled."
12817
+ language_confidence_threshold: import_zod3.z.number().nullable().describe(
12818
+ "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"
12501
12819
  ),
12502
- webhook_url: import_zod3.z.string().nullish().describe(
12503
- "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"
12504
- ),
12505
- webhook_status_code: import_zod3.z.number().nullish().describe(
12506
- "The status code we received from your server when delivering the transcript completed or failed webhook request, if a webhook URL was provided"
12507
- ),
12508
- webhook_auth: import_zod3.z.boolean().describe("Whether webhook authentication details were provided"),
12509
- webhook_auth_header_name: import_zod3.z.string().nullish().describe(
12510
- "The header name to be sent with the transcript completed or failed webhook requests"
12820
+ language_detection: import_zod3.z.boolean().nullish().describe(
12821
+ "Whether [Automatic language detection](/docs/pre-recorded-audio/automatic-language-detection) is enabled, either true or false"
12511
12822
  ),
12512
- speed_boost: import_zod3.z.boolean().nullish().describe("Whether speed boost is enabled"),
12513
- auto_highlights: import_zod3.z.boolean().describe("Whether Key Phrases is enabled, either true or false"),
12514
- auto_highlights_result: import_zod3.z.object({
12515
- status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
12516
- results: import_zod3.z.array(
12517
- import_zod3.z.object({
12518
- count: import_zod3.z.number().describe("The total number of times the key phrase appears in the audio file"),
12519
- rank: import_zod3.z.number().min(createTranscriptResponseAutoHighlightsResultResultsItemRankMin).max(createTranscriptResponseAutoHighlightsResultResultsItemRankMax).describe(
12520
- "The total relevancy to the overall audio file of this key phrase - a greater number means more relevant"
12521
- ),
12522
- text: import_zod3.z.string().describe("The text itself of the key phrase"),
12523
- timestamps: import_zod3.z.array(
12524
- import_zod3.z.object({
12525
- start: import_zod3.z.number().describe("The start time in milliseconds"),
12526
- end: import_zod3.z.number().describe("The end time in milliseconds")
12527
- }).describe("Timestamp containing a start and end property in milliseconds")
12528
- ).describe("The timestamp of the of the key phrase")
12529
- })
12530
- ).describe("A temporally-sequential array of Key Phrases")
12531
- }).describe(
12532
- "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"
12533
- ).or(import_zod3.z.null()).optional().describe(
12534
- "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"
12823
+ language_detection_options: import_zod3.z.object({
12824
+ expected_languages: import_zod3.z.array(import_zod3.z.string()).optional().describe(
12825
+ '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.'
12826
+ ),
12827
+ fallback_language: import_zod3.z.string().default(createTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault).describe(
12828
+ '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'
12829
+ ),
12830
+ code_switching: import_zod3.z.boolean().optional().describe(
12831
+ "Whether [code switching](/docs/speech-to-text/pre-recorded-audio/code-switching) should be detected.\n"
12832
+ ),
12833
+ code_switching_confidence_threshold: import_zod3.z.number().default(
12834
+ createTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault
12835
+ ).describe(
12836
+ "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"
12837
+ )
12838
+ }).optional().describe(
12839
+ "Specify options for [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection)."
12535
12840
  ),
12536
- audio_start_from: import_zod3.z.number().nullish().describe(
12537
- "The point in time, in milliseconds, in the file at which the transcription was started"
12841
+ multichannel: import_zod3.z.boolean().nullish().describe(
12842
+ "Whether [Multichannel transcription](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) was enabled in the transcription request, either true or false"
12538
12843
  ),
12539
- audio_end_at: import_zod3.z.number().nullish().describe(
12540
- "The point in time, in milliseconds, in the file at which the transcription was terminated"
12844
+ prompt: import_zod3.z.string().optional().describe(
12845
+ "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"
12541
12846
  ),
12542
- word_boost: import_zod3.z.array(import_zod3.z.string()).optional().describe("The list of custom vocabulary to boost transcription probability for"),
12543
- boost_param: import_zod3.z.string().nullish().describe("The word boost parameter value"),
12544
- filter_profanity: import_zod3.z.boolean().nullish().describe(
12545
- "Whether [Profanity Filtering](https://www.assemblyai.com/docs/models/speech-recognition#profanity-filtering) is enabled, either true or false"
12847
+ punctuate: import_zod3.z.boolean().nullish().describe(
12848
+ "Whether [Automatic Punctuation](https://www.assemblyai.com/docs/pre-recorded-audio) is enabled, either true or false"
12546
12849
  ),
12547
12850
  redact_pii: import_zod3.z.boolean().describe(
12548
- "Whether [PII Redaction](https://www.assemblyai.com/docs/models/pii-redaction) is enabled, either true or false"
12851
+ "Whether [PII Redaction](https://www.assemblyai.com/docs/pii-redaction) is enabled, either true or false"
12549
12852
  ),
12550
12853
  redact_pii_audio: import_zod3.z.boolean().nullish().describe(
12551
- "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"
12854
+ "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"
12855
+ ),
12856
+ redact_pii_audio_options: import_zod3.z.object({
12857
+ return_redacted_no_speech_audio: import_zod3.z.boolean().optional().describe(
12858
+ "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`."
12859
+ ),
12860
+ override_audio_redaction_method: import_zod3.z.enum(["silence"]).optional().describe(
12861
+ "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."
12862
+ )
12863
+ }).optional().describe(
12864
+ "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"
12552
12865
  ),
12553
12866
  redact_pii_audio_quality: import_zod3.z.enum(["mp3", "wav"]).describe(
12554
- "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."
12867
+ "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."
12555
12868
  ).or(import_zod3.z.null()).optional().describe(
12556
- "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"
12869
+ "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"
12557
12870
  ),
12558
12871
  redact_pii_policies: import_zod3.z.array(
12559
12872
  import_zod3.z.enum([
@@ -12603,19 +12916,347 @@ var createTranscriptResponse = import_zod3.z.object({
12603
12916
  "zodiac_sign"
12604
12917
  ]).describe("The type of PII to redact")
12605
12918
  ).nullish().describe(
12606
- "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"
12919
+ "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"
12607
12920
  ),
12608
12921
  redact_pii_sub: import_zod3.z.enum(["entity_name", "hash"]).optional().describe(
12609
- '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.'
12922
+ "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."
12923
+ ),
12924
+ sentiment_analysis: import_zod3.z.boolean().nullish().describe(
12925
+ "Whether [Sentiment Analysis](https://www.assemblyai.com/docs/speech-understanding/sentiment-analysis) is enabled, can be true or false"
12926
+ ),
12927
+ sentiment_analysis_results: import_zod3.z.array(
12928
+ import_zod3.z.object({
12929
+ text: import_zod3.z.string().describe("The transcript of the sentence"),
12930
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, of the sentence"),
12931
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, of the sentence"),
12932
+ sentiment: import_zod3.z.enum(["POSITIVE", "NEUTRAL", "NEGATIVE"]),
12933
+ confidence: import_zod3.z.number().describe(
12934
+ "The confidence score for the detected sentiment of the sentence, from 0 to 1"
12935
+ ),
12936
+ channel: import_zod3.z.string().nullish().describe(
12937
+ "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
12938
+ ),
12939
+ speaker: import_zod3.z.string().nullable().describe(
12940
+ "The speaker of the sentence if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
12941
+ )
12942
+ }).describe("The result of the Sentiment Analysis model")
12943
+ ).nullish().describe(
12944
+ "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"
12610
12945
  ),
12611
12946
  speaker_labels: import_zod3.z.boolean().nullish().describe(
12612
- "Whether [Speaker diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, can be true or false"
12947
+ "Whether [Speaker diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, can be true or false"
12613
12948
  ),
12614
12949
  speakers_expected: import_zod3.z.number().nullish().describe(
12615
- "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."
12950
+ "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."
12951
+ ),
12952
+ speech_model_used: import_zod3.z.string().optional().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
+ speech_models: import_zod3.z.array(
12956
+ import_zod3.z.string().describe(
12957
+ "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."
12958
+ )
12959
+ ).nullish().describe(
12960
+ "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"
12961
+ ),
12962
+ speech_threshold: import_zod3.z.number().nullish().describe(
12963
+ "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"
12964
+ ),
12965
+ speech_understanding: import_zod3.z.object({
12966
+ request: import_zod3.z.object({
12967
+ translation: import_zod3.z.object({
12968
+ target_languages: import_zod3.z.array(import_zod3.z.string()).describe(
12969
+ 'List of target language codes (e.g., `["es", "de"]`). See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for supported languages.'
12970
+ ),
12971
+ formal: import_zod3.z.boolean().default(createTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault).describe(
12972
+ "Use formal language style. See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for more details."
12973
+ ),
12974
+ match_original_utterance: import_zod3.z.boolean().optional().describe(
12975
+ "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."
12976
+ )
12977
+ })
12978
+ }).describe(
12979
+ "Request body for [Translation](https://www.assemblyai.com/docs/speech-understanding/translation)."
12980
+ ).or(
12981
+ import_zod3.z.object({
12982
+ speaker_identification: import_zod3.z.object({
12983
+ speaker_type: import_zod3.z.enum(["role", "name"]).describe(
12984
+ "Type of speaker identification. See [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification) for details on each type."
12985
+ ),
12986
+ known_values: import_zod3.z.array(import_zod3.z.string()).optional().describe(
12987
+ 'Required if speaker_type is "role". Each value must be 35 characters or less.'
12988
+ ),
12989
+ speakers: import_zod3.z.array(
12990
+ import_zod3.z.object({
12991
+ role: import_zod3.z.string().optional().describe(
12992
+ 'The role of the speaker. Required when `speaker_type` is "role".'
12993
+ ),
12994
+ name: import_zod3.z.string().optional().describe(
12995
+ 'The name of the speaker. Required when `speaker_type` is "name".'
12996
+ ),
12997
+ description: import_zod3.z.string().optional().describe(
12998
+ "A description of the speaker to help the model identify them based on conversational context."
12999
+ )
13000
+ })
13001
+ ).optional().describe(
13002
+ "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."
13003
+ )
13004
+ })
13005
+ }).describe(
13006
+ "Request body for [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification)."
13007
+ )
13008
+ ).or(
13009
+ import_zod3.z.object({
13010
+ custom_formatting: import_zod3.z.object({
13011
+ date: import_zod3.z.string().optional().describe(
13012
+ 'Date format pattern (e.g., `"mm/dd/yyyy"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
13013
+ ),
13014
+ phone_number: import_zod3.z.string().optional().describe(
13015
+ 'Phone number format pattern (e.g., `"(xxx)xxx-xxxx"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
13016
+ ),
13017
+ email: import_zod3.z.string().optional().describe(
13018
+ 'Email format pattern (e.g., `"username@domain.com"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
13019
+ )
13020
+ })
13021
+ }).describe(
13022
+ "Request body for [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting)."
13023
+ )
13024
+ ).optional(),
13025
+ response: import_zod3.z.object({
13026
+ translation: import_zod3.z.object({
13027
+ status: import_zod3.z.string().optional()
13028
+ }).optional()
13029
+ }).or(
13030
+ import_zod3.z.object({
13031
+ speaker_identification: import_zod3.z.object({
13032
+ mapping: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional().describe(
13033
+ 'A mapping of the original generic speaker labels (e.g., "A", "B") to the identified speaker names or roles.'
13034
+ ),
13035
+ status: import_zod3.z.string().optional()
13036
+ }).optional()
13037
+ })
13038
+ ).or(
13039
+ import_zod3.z.object({
13040
+ custom_formatting: import_zod3.z.object({
13041
+ mapping: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional(),
13042
+ formatted_text: import_zod3.z.string().optional()
13043
+ }).optional()
13044
+ })
13045
+ ).optional()
13046
+ }).optional().describe(
13047
+ "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"
13048
+ ),
13049
+ status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).describe(
13050
+ "The status of your transcript. Possible values are queued, processing, completed, or error."
13051
+ ),
13052
+ summarization: import_zod3.z.boolean().describe(
13053
+ "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"
13054
+ ),
13055
+ summary: import_zod3.z.string().nullish().describe(
13056
+ "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."
13057
+ ),
13058
+ summary_model: import_zod3.z.string().nullish().describe(
13059
+ "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"
13060
+ ),
13061
+ summary_type: import_zod3.z.string().nullish().describe(
13062
+ "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."
13063
+ ),
13064
+ remove_audio_tags: import_zod3.z.enum(["all"]).describe(
13065
+ "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"
13066
+ ).or(import_zod3.z.null()).optional().describe(
13067
+ "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"
13068
+ ),
13069
+ temperature: import_zod3.z.number().nullish().describe(
13070
+ "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"
13071
+ ),
13072
+ text: import_zod3.z.string().nullish().describe("The textual transcript of your media file"),
13073
+ throttled: import_zod3.z.boolean().nullish().describe(
13074
+ "True while a request is throttled and false when a request is no longer throttled"
13075
+ ),
13076
+ utterances: import_zod3.z.array(
13077
+ import_zod3.z.object({
13078
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this utterance"),
13079
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, of the utterance in the audio file"),
13080
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, of the utterance in the audio file"),
13081
+ text: import_zod3.z.string().describe("The text for this utterance"),
13082
+ words: import_zod3.z.array(
13083
+ import_zod3.z.object({
13084
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
13085
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
13086
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
13087
+ text: import_zod3.z.string().describe("The text of the word"),
13088
+ channel: import_zod3.z.string().nullish().describe(
13089
+ "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13090
+ ),
13091
+ speaker: import_zod3.z.string().nullable().describe(
13092
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
13093
+ )
13094
+ })
13095
+ ).describe("The words in the utterance."),
13096
+ channel: import_zod3.z.string().nullish().describe(
13097
+ "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13098
+ ),
13099
+ speaker: import_zod3.z.string().describe(
13100
+ '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.'
13101
+ ),
13102
+ translated_texts: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional().describe(
13103
+ 'Translations keyed by language code (e.g., `{"es": "Texto traducido", "de": "\xDCbersetzter Text"}`). Only present when `match_original_utterance` is enabled with translation.'
13104
+ )
13105
+ })
13106
+ ).nullish().describe(
13107
+ "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"
13108
+ ),
13109
+ webhook_auth: import_zod3.z.boolean().describe(
13110
+ "Whether [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) authentication details were provided"
13111
+ ),
13112
+ webhook_auth_header_name: import_zod3.z.string().nullish().describe(
13113
+ "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"
13114
+ ),
13115
+ webhook_status_code: import_zod3.z.number().nullish().describe(
13116
+ "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"
13117
+ ),
13118
+ webhook_url: import_zod3.z.string().nullish().describe(
13119
+ "The URL to which we send [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) requests.\n"
13120
+ ),
13121
+ words: import_zod3.z.array(
13122
+ import_zod3.z.object({
13123
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
13124
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
13125
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
13126
+ text: import_zod3.z.string().describe("The text of the word"),
13127
+ channel: import_zod3.z.string().nullish().describe(
13128
+ "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13129
+ ),
13130
+ speaker: import_zod3.z.string().nullable().describe(
13131
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
13132
+ )
13133
+ })
13134
+ ).nullish().describe(
13135
+ "An array of temporally-sequential word objects, one for each word in the transcript.\n"
13136
+ ),
13137
+ acoustic_model: import_zod3.z.string().describe("This parameter does not currently have any functionality attached to it."),
13138
+ custom_topics: import_zod3.z.boolean().nullish().describe("This parameter does not currently have any functionality attached to it."),
13139
+ language_model: import_zod3.z.string().describe("This parameter does not currently have any functionality attached to it."),
13140
+ speech_model: import_zod3.z.string().describe(
13141
+ "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."
13142
+ ).or(import_zod3.z.null()).describe(
13143
+ "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"
13144
+ ),
13145
+ speed_boost: import_zod3.z.boolean().nullish().describe("This parameter does not currently have any functionality attached to it."),
13146
+ topics: import_zod3.z.array(import_zod3.z.string()).optional().describe("This parameter does not currently have any functionality attached to it."),
13147
+ translated_texts: import_zod3.z.object({
13148
+ language_code: import_zod3.z.string().optional().describe("Translated text for this language code")
13149
+ }).optional().describe(
13150
+ "Translated text keyed by language code. See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for more details."
13151
+ )
13152
+ }).describe("A transcript object");
13153
+ var listTranscriptsQueryLimitDefault = 10;
13154
+ var listTranscriptsQueryThrottledOnlyDefault = false;
13155
+ var listTranscriptsQueryParams = import_zod3.z.object({
13156
+ limit: import_zod3.z.number().default(listTranscriptsQueryLimitDefault).describe("Maximum amount of transcripts to retrieve"),
13157
+ status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).optional().describe("Filter by transcript status"),
13158
+ created_on: import_zod3.z.string().date().optional().describe("Only get transcripts created on this date"),
13159
+ before_id: import_zod3.z.string().uuid().optional().describe("Get transcripts that were created before this transcript ID"),
13160
+ after_id: import_zod3.z.string().uuid().optional().describe("Get transcripts that were created after this transcript ID"),
13161
+ throttled_only: import_zod3.z.boolean().optional().describe("Only get throttled transcripts, overrides the status filter")
13162
+ });
13163
+ var listTranscriptsResponse = import_zod3.z.object({
13164
+ page_details: import_zod3.z.object({
13165
+ limit: import_zod3.z.number().describe("The number of results this page is limited to"),
13166
+ result_count: import_zod3.z.number().describe("The actual number of results in the page"),
13167
+ current_url: import_zod3.z.string().describe("The URL used to retrieve the current page of transcripts"),
13168
+ prev_url: import_zod3.z.string().nullable().describe(
13169
+ "The URL to the next page of transcripts. The previous URL always points to a page with older transcripts."
13170
+ ),
13171
+ next_url: import_zod3.z.string().nullable().describe(
13172
+ "The URL to the next page of transcripts. The next URL always points to a page with newer transcripts."
13173
+ )
13174
+ }).describe(
13175
+ "Details of the transcript page. Transcripts are sorted from newest to oldest. The previous URL always points to a page with older transcripts."
13176
+ ),
13177
+ transcripts: import_zod3.z.array(
13178
+ import_zod3.z.object({
13179
+ id: import_zod3.z.string().uuid().describe("The unique identifier for the transcript"),
13180
+ resource_url: import_zod3.z.string().describe("The URL to retrieve the transcript"),
13181
+ status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).describe(
13182
+ "The status of your transcript. Possible values are queued, processing, completed, or error."
13183
+ ),
13184
+ created: import_zod3.z.string().datetime({}).describe("The date and time the transcript was created"),
13185
+ completed: import_zod3.z.string().datetime({}).optional().describe("The date and time the transcript was completed"),
13186
+ audio_url: import_zod3.z.string().describe("The URL to the audio file"),
13187
+ error: import_zod3.z.string().nullable().describe("Error message of why the transcript failed")
13188
+ })
13189
+ ).describe("An array of transcripts")
13190
+ }).describe(
13191
+ "A list of transcripts. Transcripts are sorted from newest to oldest. The previous URL always points to a page with older transcripts."
13192
+ );
13193
+ var getTranscriptParams = import_zod3.z.object({
13194
+ transcript_id: import_zod3.z.string().describe("ID of the transcript")
13195
+ });
13196
+ var getTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault = "auto";
13197
+ var getTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault = false;
13198
+ var getTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault = 0.3;
13199
+ var getTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault = false;
13200
+ var getTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault = true;
13201
+ var getTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault = false;
13202
+ var getTranscriptResponse = import_zod3.z.object({
13203
+ audio_channels: import_zod3.z.number().optional().describe(
13204
+ "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."
13205
+ ),
13206
+ audio_duration: import_zod3.z.number().nullish().describe("The duration of this transcript object's media file, in seconds"),
13207
+ audio_end_at: import_zod3.z.number().nullish().describe(
13208
+ "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."
13209
+ ),
13210
+ audio_start_from: import_zod3.z.number().nullish().describe(
13211
+ "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."
13212
+ ),
13213
+ audio_url: import_zod3.z.string().describe("The URL of the media that was transcribed"),
13214
+ auto_chapters: import_zod3.z.boolean().nullish().describe(
13215
+ "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"
13216
+ ),
13217
+ auto_highlights: import_zod3.z.boolean().describe(
13218
+ "Whether [Key Phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases) is enabled, either true or false"
13219
+ ),
13220
+ auto_highlights_result: import_zod3.z.object({
13221
+ status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
13222
+ results: import_zod3.z.array(
13223
+ import_zod3.z.object({
13224
+ count: import_zod3.z.number().describe("The total number of times the key phrase appears in the audio file"),
13225
+ rank: import_zod3.z.number().describe(
13226
+ "The total relevancy to the overall audio file of this key phrase - a greater number means more relevant"
13227
+ ),
13228
+ text: import_zod3.z.string().describe("The text itself of the key phrase"),
13229
+ timestamps: import_zod3.z.array(
13230
+ import_zod3.z.object({
13231
+ start: import_zod3.z.number().describe("The start time in milliseconds"),
13232
+ end: import_zod3.z.number().describe("The end time in milliseconds")
13233
+ }).describe("Timestamp containing a start and end property in milliseconds")
13234
+ ).describe("The timestamp of the of the key phrase")
13235
+ })
13236
+ ).describe("A temporally-sequential array of Key Phrases")
13237
+ }).describe(
13238
+ "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"
13239
+ ).or(import_zod3.z.null()).optional().describe(
13240
+ "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"
13241
+ ),
13242
+ chapters: import_zod3.z.array(
13243
+ import_zod3.z.object({
13244
+ gist: import_zod3.z.string().describe(
13245
+ "An ultra-short summary (just a few words) of the content spoken in the chapter"
13246
+ ),
13247
+ headline: import_zod3.z.string().describe("A single sentence summary of the content spoken during the chapter"),
13248
+ summary: import_zod3.z.string().describe("A one paragraph summary of the content spoken during the chapter"),
13249
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter"),
13250
+ end: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter")
13251
+ }).describe("Chapter of the audio file")
13252
+ ).nullish().describe(
13253
+ "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."
13254
+ ),
13255
+ confidence: import_zod3.z.number().nullish().describe(
13256
+ "The confidence score for the transcript, between 0.0 (low confidence) and 1.0 (high confidence)"
12616
13257
  ),
12617
13258
  content_safety: import_zod3.z.boolean().nullish().describe(
12618
- "Whether [Content Moderation](https://www.assemblyai.com/docs/models/content-moderation) is enabled, can be true or false"
13259
+ "Whether [Content Moderation](https://www.assemblyai.com/docs/content-moderation) is enabled, can be true or false"
12619
13260
  ),
12620
13261
  content_safety_labels: import_zod3.z.object({
12621
13262
  status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
@@ -12625,16 +13266,8 @@ var createTranscriptResponse = import_zod3.z.object({
12625
13266
  labels: import_zod3.z.array(
12626
13267
  import_zod3.z.object({
12627
13268
  label: import_zod3.z.string().describe("The label of the sensitive topic"),
12628
- confidence: import_zod3.z.number().min(
12629
- createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin
12630
- ).max(
12631
- createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax
12632
- ).describe("The confidence score for the topic being discussed, from 0 to 1"),
12633
- severity: import_zod3.z.number().min(
12634
- createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin
12635
- ).max(
12636
- createTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax
12637
- ).describe("How severely the topic is discussed in the section, from 0 to 1")
13269
+ confidence: import_zod3.z.number().describe("The confidence score for the topic being discussed, from 0 to 1"),
13270
+ severity: import_zod3.z.number().describe("How severely the topic is discussed in the section, from 0 to 1")
12638
13271
  })
12639
13272
  ).describe(
12640
13273
  "An array of safety labels, one per sensitive topic that was detected in the section"
@@ -12647,128 +13280,39 @@ var createTranscriptResponse = import_zod3.z.object({
12647
13280
  }).describe("Timestamp containing a start and end property in milliseconds")
12648
13281
  })
12649
13282
  ).describe("An array of results for the Content Moderation model"),
12650
- summary: import_zod3.z.record(
12651
- import_zod3.z.string(),
12652
- import_zod3.z.number().min(createTranscriptResponseContentSafetyLabelsSummaryMinOne).max(createTranscriptResponseContentSafetyLabelsSummaryMaxOne).describe(
12653
- 'A confidence score for the presence of the sensitive topic "topic" across the entire audio file'
12654
- )
12655
- ).describe(
13283
+ summary: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.number()).describe(
12656
13284
  "A summary of the Content Moderation confidence results for the entire audio file"
12657
13285
  ),
12658
13286
  severity_score_summary: import_zod3.z.record(
12659
13287
  import_zod3.z.string(),
12660
13288
  import_zod3.z.object({
12661
- low: import_zod3.z.number().min(createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin).max(createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax),
12662
- medium: import_zod3.z.number().min(createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin).max(createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax),
12663
- high: import_zod3.z.number().min(createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin).max(createTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax)
13289
+ low: import_zod3.z.number(),
13290
+ medium: import_zod3.z.number(),
13291
+ high: import_zod3.z.number()
12664
13292
  })
12665
13293
  ).describe(
12666
13294
  "A summary of the Content Moderation severity results for the entire audio file"
12667
13295
  )
12668
13296
  }).describe(
12669
- "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"
13297
+ "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"
12670
13298
  ).or(import_zod3.z.null()).optional().describe(
12671
- "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"
12672
- ),
12673
- iab_categories: import_zod3.z.boolean().nullish().describe(
12674
- "Whether [Topic Detection](https://www.assemblyai.com/docs/models/topic-detection) is enabled, can be true or false"
12675
- ),
12676
- iab_categories_result: import_zod3.z.object({
12677
- status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
12678
- results: import_zod3.z.array(
12679
- import_zod3.z.object({
12680
- text: import_zod3.z.string().describe("The text in the transcript in which a detected topic occurs"),
12681
- labels: import_zod3.z.array(
12682
- import_zod3.z.object({
12683
- relevance: import_zod3.z.number().min(
12684
- createTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin
12685
- ).max(
12686
- createTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax
12687
- ).describe("How relevant the detected topic is of a detected topic"),
12688
- label: import_zod3.z.string().describe(
12689
- "The IAB taxonomical label for the label of the detected topic, where > denotes supertopic/subtopic relationship"
12690
- )
12691
- })
12692
- ).optional().describe("An array of detected topics in the text"),
12693
- timestamp: import_zod3.z.object({
12694
- start: import_zod3.z.number().describe("The start time in milliseconds"),
12695
- end: import_zod3.z.number().describe("The end time in milliseconds")
12696
- }).optional().describe("Timestamp containing a start and end property in milliseconds")
12697
- }).describe("The result of the topic detection model")
12698
- ).describe("An array of results for the Topic Detection model"),
12699
- summary: import_zod3.z.record(
12700
- import_zod3.z.string(),
12701
- import_zod3.z.number().min(createTranscriptResponseIabCategoriesResultSummaryMinOne).max(createTranscriptResponseIabCategoriesResultSummaryMaxOne)
12702
- ).describe("The overall relevance of topic to the entire audio file")
12703
- }).describe(
12704
- "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"
12705
- ).or(import_zod3.z.null()).optional().describe(
12706
- "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"
13299
+ "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"
12707
13300
  ),
12708
13301
  custom_spelling: import_zod3.z.array(
12709
13302
  import_zod3.z.object({
12710
- from: import_zod3.z.array(import_zod3.z.string().describe("Word or phrase to replace")).describe("Words or phrases to replace"),
13303
+ from: import_zod3.z.array(import_zod3.z.string()).describe("Words or phrases to replace"),
12711
13304
  to: import_zod3.z.string().describe("Word to replace with")
12712
13305
  }).describe(
12713
13306
  "Object containing words or phrases to replace, and the word or phrase to replace with"
12714
13307
  )
12715
- ).nullish().describe("Customize how words are spelled and formatted using to and from values"),
12716
- keyterms_prompt: import_zod3.z.array(import_zod3.z.string()).optional().describe(
12717
- "Improve accuracy with up to 1000 domain-specific words or phrases (maximum 6 words per phrase).\n"
12718
- ),
12719
- prompt: import_zod3.z.string().optional().describe("This parameter does not currently have any functionality attached to it."),
12720
- auto_chapters: import_zod3.z.boolean().nullish().describe(
12721
- "Whether [Auto Chapters](https://www.assemblyai.com/docs/models/auto-chapters) is enabled, can be true or false"
12722
- ),
12723
- chapters: import_zod3.z.array(
12724
- import_zod3.z.object({
12725
- gist: import_zod3.z.string().describe(
12726
- "An ultra-short summary (just a few words) of the content spoken in the chapter"
12727
- ),
12728
- headline: import_zod3.z.string().describe("A single sentence summary of the content spoken during the chapter"),
12729
- summary: import_zod3.z.string().describe("A one paragraph summary of the content spoken during the chapter"),
12730
- start: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter"),
12731
- end: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter")
12732
- }).describe("Chapter of the audio file")
12733
- ).nullish().describe("An array of temporally sequential chapters for the audio file"),
12734
- summarization: import_zod3.z.boolean().describe(
12735
- "Whether [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled, either true or false"
12736
- ),
12737
- summary_type: import_zod3.z.string().nullish().describe(
12738
- "The type of summary generated, if [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled"
12739
- ),
12740
- summary_model: import_zod3.z.string().nullish().describe(
12741
- "The Summarization model used to generate the summary,\nif [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled\n"
12742
- ),
12743
- summary: import_zod3.z.string().nullish().describe(
12744
- "The generated summary of the media file, if [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled"
12745
- ),
12746
- custom_topics: import_zod3.z.boolean().nullish().describe("Whether custom topics is enabled, either true or false"),
12747
- topics: import_zod3.z.array(import_zod3.z.string()).optional().describe("The list of custom topics provided if custom topics is enabled"),
12748
- sentiment_analysis: import_zod3.z.boolean().nullish().describe(
12749
- "Whether [Sentiment Analysis](https://www.assemblyai.com/docs/models/sentiment-analysis) is enabled, can be true or false"
12750
- ),
12751
- sentiment_analysis_results: import_zod3.z.array(
12752
- import_zod3.z.object({
12753
- text: import_zod3.z.string().describe("The transcript of the sentence"),
12754
- start: import_zod3.z.number().describe("The starting time, in milliseconds, of the sentence"),
12755
- end: import_zod3.z.number().describe("The ending time, in milliseconds, of the sentence"),
12756
- sentiment: import_zod3.z.enum(["POSITIVE", "NEUTRAL", "NEGATIVE"]),
12757
- confidence: import_zod3.z.number().min(createTranscriptResponseSentimentAnalysisResultsItemConfidenceMin).max(createTranscriptResponseSentimentAnalysisResultsItemConfidenceMax).describe(
12758
- "The confidence score for the detected sentiment of the sentence, from 0 to 1"
12759
- ),
12760
- channel: import_zod3.z.string().nullish().describe(
12761
- "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
12762
- ),
12763
- speaker: import_zod3.z.string().nullable().describe(
12764
- "The speaker of the sentence if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
12765
- )
12766
- }).describe("The result of the Sentiment Analysis model")
12767
13308
  ).nullish().describe(
12768
- "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"
13309
+ "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."
12769
13310
  ),
12770
- entity_detection: import_zod3.z.boolean().nullish().describe(
12771
- "Whether [Entity Detection](https://www.assemblyai.com/docs/models/entity-detection) is enabled, can be true or false"
13311
+ disfluencies: import_zod3.z.boolean().nullish().describe(
13312
+ 'Transcribe [Filler Words](https://www.assemblyai.com/docs/pre-recorded-audio/filler-words), like "umm", in your media file; can be true or false'
13313
+ ),
13314
+ domain: import_zod3.z.string().nullish().describe(
13315
+ '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'
12772
13316
  ),
12773
13317
  entities: import_zod3.z.array(
12774
13318
  import_zod3.z.object({
@@ -12827,108 +13371,49 @@ var createTranscriptResponse = import_zod3.z.object({
12827
13371
  )
12828
13372
  }).describe("A detected entity")
12829
13373
  ).nullish().describe(
12830
- "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"
12831
- ),
12832
- speech_threshold: import_zod3.z.number().min(createTranscriptResponseSpeechThresholdMin).max(createTranscriptResponseSpeechThresholdMax).nullish().describe(
12833
- "Defaults to null. Reject audio files that contain less than this fraction of speech.\nValid values are in the range [0, 1] inclusive.\n"
13374
+ "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"
12834
13375
  ),
12835
- throttled: import_zod3.z.boolean().nullish().describe(
12836
- "True while a request is throttled and false when a request is no longer throttled"
13376
+ entity_detection: import_zod3.z.boolean().nullish().describe(
13377
+ "Whether [Entity Detection](https://www.assemblyai.com/docs/speech-understanding/entity-detection) is enabled, can be true or false"
12837
13378
  ),
12838
13379
  error: import_zod3.z.string().optional().describe("Error message of why the transcript failed"),
12839
- language_model: import_zod3.z.string().describe("The language model that was used for the transcript"),
12840
- acoustic_model: import_zod3.z.string().describe("The acoustic model that was used for the transcript")
12841
- }).describe("A transcript object");
12842
- var listTranscriptsQueryLimitDefault = 10;
12843
- var listTranscriptsQueryLimitMax = 200;
12844
- var listTranscriptsQueryThrottledOnlyDefault = false;
12845
- var listTranscriptsQueryParams = import_zod3.z.object({
12846
- limit: import_zod3.z.number().min(1).max(listTranscriptsQueryLimitMax).default(listTranscriptsQueryLimitDefault).describe("Maximum amount of transcripts to retrieve"),
12847
- status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).optional().describe("Filter by transcript status"),
12848
- created_on: import_zod3.z.string().date().optional().describe("Only get transcripts created on this date"),
12849
- before_id: import_zod3.z.string().uuid().optional().describe("Get transcripts that were created before this transcript ID"),
12850
- after_id: import_zod3.z.string().uuid().optional().describe("Get transcripts that were created after this transcript ID"),
12851
- throttled_only: import_zod3.z.boolean().optional().describe("Only get throttled transcripts, overrides the status filter")
12852
- });
12853
- var listTranscriptsResponseTranscriptsItemCreatedRegExp = new RegExp(
12854
- "^(?:(\\d{4}-\\d{2}-\\d{2})T(\\d{2}:\\d{2}:\\d{2}(?:\\.\\d+)?))$"
12855
- );
12856
- var listTranscriptsResponseTranscriptsItemCompletedRegExp = new RegExp(
12857
- "^(?:(\\d{4}-\\d{2}-\\d{2})T(\\d{2}:\\d{2}:\\d{2}(?:\\.\\d+)?))$"
12858
- );
12859
- var listTranscriptsResponse = import_zod3.z.object({
12860
- page_details: import_zod3.z.object({
12861
- limit: import_zod3.z.number().describe("The number of results this page is limited to"),
12862
- result_count: import_zod3.z.number().describe("The actual number of results in the page"),
12863
- current_url: import_zod3.z.string().describe("The URL used to retrieve the current page of transcripts"),
12864
- prev_url: import_zod3.z.string().nullable().describe(
12865
- "The URL to the next page of transcripts. The previous URL always points to a page with older transcripts."
12866
- ),
12867
- next_url: import_zod3.z.string().nullable().describe(
12868
- "The URL to the next page of transcripts. The next URL always points to a page with newer transcripts."
12869
- )
13380
+ filter_profanity: import_zod3.z.boolean().nullish().describe(
13381
+ "Whether [Profanity Filtering](https://www.assemblyai.com/docs/profanity-filtering) is enabled, either true or false"
13382
+ ),
13383
+ format_text: import_zod3.z.boolean().nullish().describe(
13384
+ "Whether [Text Formatting](https://www.assemblyai.com/docs/pre-recorded-audio) is enabled, either true or false"
13385
+ ),
13386
+ iab_categories: import_zod3.z.boolean().nullish().describe(
13387
+ "Whether [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection) is enabled, can be true or false"
13388
+ ),
13389
+ iab_categories_result: import_zod3.z.object({
13390
+ status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
13391
+ results: import_zod3.z.array(
13392
+ import_zod3.z.object({
13393
+ text: import_zod3.z.string().describe("The text in the transcript in which a detected topic occurs"),
13394
+ labels: import_zod3.z.array(
13395
+ import_zod3.z.object({
13396
+ relevance: import_zod3.z.number().describe("How relevant the detected topic is of a detected topic"),
13397
+ label: import_zod3.z.string().describe(
13398
+ "The IAB taxonomical label for the label of the detected topic, where > denotes supertopic/subtopic relationship"
13399
+ )
13400
+ })
13401
+ ).optional().describe("An array of detected topics in the text"),
13402
+ timestamp: import_zod3.z.object({
13403
+ start: import_zod3.z.number().describe("The start time in milliseconds"),
13404
+ end: import_zod3.z.number().describe("The end time in milliseconds")
13405
+ }).optional().describe("Timestamp containing a start and end property in milliseconds")
13406
+ }).describe("The result of the topic detection model")
13407
+ ).describe("An array of results for the Topic Detection model"),
13408
+ summary: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.number()).describe("The overall relevance of topic to the entire audio file")
12870
13409
  }).describe(
12871
- "Details of the transcript page. Transcripts are sorted from newest to oldest. The previous URL always points to a page with older transcripts."
13410
+ "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"
13411
+ ).or(import_zod3.z.null()).optional().describe(
13412
+ "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"
12872
13413
  ),
12873
- transcripts: import_zod3.z.array(
12874
- import_zod3.z.object({
12875
- id: import_zod3.z.string().uuid().describe("The unique identifier for the transcript"),
12876
- resource_url: import_zod3.z.string().describe("The URL to retrieve the transcript"),
12877
- status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).describe(
12878
- "The status of your transcript. Possible values are queued, processing, completed, or error."
12879
- ),
12880
- created: import_zod3.z.string().regex(listTranscriptsResponseTranscriptsItemCreatedRegExp).describe("The date and time the transcript was created"),
12881
- completed: import_zod3.z.string().regex(listTranscriptsResponseTranscriptsItemCompletedRegExp).nullable().describe("The date and time the transcript was completed"),
12882
- audio_url: import_zod3.z.string().describe("The URL to the audio file"),
12883
- error: import_zod3.z.string().nullable().describe("Error message of why the transcript failed")
12884
- })
12885
- ).describe("An array of transcripts")
12886
- }).describe(
12887
- "A list of transcripts. Transcripts are sorted from newest to oldest. The previous URL always points to a page with older transcripts."
12888
- );
12889
- var getTranscriptParams = import_zod3.z.object({
12890
- transcript_id: import_zod3.z.string().describe("ID of the transcript")
12891
- });
12892
- var getTranscriptResponseLanguageConfidenceThresholdMin = 0;
12893
- var getTranscriptResponseLanguageConfidenceThresholdMax = 1;
12894
- var getTranscriptResponseLanguageConfidenceMin = 0;
12895
- var getTranscriptResponseLanguageConfidenceMax = 1;
12896
- var getTranscriptResponseSpeechModelDefault = null;
12897
- var getTranscriptResponseWordsItemConfidenceMin = 0;
12898
- var getTranscriptResponseWordsItemConfidenceMax = 1;
12899
- var getTranscriptResponseUtterancesItemConfidenceMin = 0;
12900
- var getTranscriptResponseUtterancesItemConfidenceMax = 1;
12901
- var getTranscriptResponseUtterancesItemWordsItemConfidenceMin = 0;
12902
- var getTranscriptResponseUtterancesItemWordsItemConfidenceMax = 1;
12903
- var getTranscriptResponseConfidenceMin = 0;
12904
- var getTranscriptResponseConfidenceMax = 1;
12905
- var getTranscriptResponseAutoHighlightsResultResultsItemRankMin = 0;
12906
- var getTranscriptResponseAutoHighlightsResultResultsItemRankMax = 1;
12907
- var getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin = 0;
12908
- var getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax = 1;
12909
- var getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin = 0;
12910
- var getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax = 1;
12911
- var getTranscriptResponseContentSafetyLabelsSummaryMinOne = 0;
12912
- var getTranscriptResponseContentSafetyLabelsSummaryMaxOne = 1;
12913
- var getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin = 0;
12914
- var getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax = 1;
12915
- var getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin = 0;
12916
- var getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax = 1;
12917
- var getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin = 0;
12918
- var getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax = 1;
12919
- var getTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin = 0;
12920
- var getTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax = 1;
12921
- var getTranscriptResponseIabCategoriesResultSummaryMinOne = 0;
12922
- var getTranscriptResponseIabCategoriesResultSummaryMaxOne = 1;
12923
- var getTranscriptResponseSentimentAnalysisResultsItemConfidenceMin = 0;
12924
- var getTranscriptResponseSentimentAnalysisResultsItemConfidenceMax = 1;
12925
- var getTranscriptResponseSpeechThresholdMin = 0;
12926
- var getTranscriptResponseSpeechThresholdMax = 1;
12927
- var getTranscriptResponse = import_zod3.z.object({
12928
13414
  id: import_zod3.z.string().uuid().describe("The unique identifier of your transcript"),
12929
- audio_url: import_zod3.z.string().describe("The URL of the media that was transcribed"),
12930
- status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).describe(
12931
- "The status of your transcript. Possible values are queued, processing, completed, or error."
13415
+ keyterms_prompt: import_zod3.z.array(import_zod3.z.string()).optional().describe(
13416
+ "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"
12932
13417
  ),
12933
13418
  language_code: import_zod3.z.enum([
12934
13419
  "en",
@@ -13033,138 +13518,175 @@ var getTranscriptResponse = import_zod3.z.object({
13033
13518
  "cy",
13034
13519
  "yi",
13035
13520
  "yo"
13036
- ]).describe(
13037
- "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"
13038
- ).or(import_zod3.z.string()).optional().describe(
13039
- "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"
13040
- ),
13041
- language_detection: import_zod3.z.boolean().nullish().describe(
13042
- "Whether [Automatic language detection](https://www.assemblyai.com/docs/models/speech-recognition#automatic-language-detection) is enabled, either true or false"
13043
- ),
13044
- language_confidence_threshold: import_zod3.z.number().min(getTranscriptResponseLanguageConfidenceThresholdMin).max(getTranscriptResponseLanguageConfidenceThresholdMax).nullable().describe(
13045
- "The confidence threshold for the automatically detected language.\nAn error will be returned if the language confidence is below this threshold.\n"
13046
- ),
13047
- language_confidence: import_zod3.z.number().min(getTranscriptResponseLanguageConfidenceMin).max(getTranscriptResponseLanguageConfidenceMax).nullable().describe(
13048
- "The confidence score for the detected language, between 0.0 (low confidence) and 1.0 (high confidence)"
13049
- ),
13050
- 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(
13051
- "The speech model used for the transcription. When `null`, the default model is used."
13052
- ),
13053
- text: import_zod3.z.string().nullish().describe("The textual transcript of your media file"),
13054
- words: import_zod3.z.array(
13055
- import_zod3.z.object({
13056
- confidence: import_zod3.z.number().min(getTranscriptResponseWordsItemConfidenceMin).max(getTranscriptResponseWordsItemConfidenceMax).describe("The confidence score for the transcript of this word"),
13057
- start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
13058
- end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
13059
- text: import_zod3.z.string().describe("The text of the word"),
13060
- channel: import_zod3.z.string().nullish().describe(
13061
- "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13062
- ),
13063
- speaker: import_zod3.z.string().nullable().describe(
13064
- "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
13065
- )
13066
- })
13067
- ).nullish().describe(
13068
- "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"
13521
+ ]).optional().describe(
13522
+ "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"
13069
13523
  ),
13070
- utterances: import_zod3.z.array(
13071
- import_zod3.z.object({
13072
- confidence: import_zod3.z.number().min(getTranscriptResponseUtterancesItemConfidenceMin).max(getTranscriptResponseUtterancesItemConfidenceMax).describe("The confidence score for the transcript of this utterance"),
13073
- start: import_zod3.z.number().describe("The starting time, in milliseconds, of the utterance in the audio file"),
13074
- end: import_zod3.z.number().describe("The ending time, in milliseconds, of the utterance in the audio file"),
13075
- text: import_zod3.z.string().describe("The text for this utterance"),
13076
- words: import_zod3.z.array(
13077
- import_zod3.z.object({
13078
- confidence: import_zod3.z.number().min(getTranscriptResponseUtterancesItemWordsItemConfidenceMin).max(getTranscriptResponseUtterancesItemWordsItemConfidenceMax).describe("The confidence score for the transcript of this word"),
13079
- start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
13080
- end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
13081
- text: import_zod3.z.string().describe("The text of the word"),
13082
- channel: import_zod3.z.string().nullish().describe(
13083
- "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13084
- ),
13085
- speaker: import_zod3.z.string().nullable().describe(
13086
- "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
13087
- )
13088
- })
13089
- ).describe("The words in the utterance."),
13090
- channel: import_zod3.z.string().nullish().describe(
13091
- "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13092
- ),
13093
- speaker: import_zod3.z.string().describe(
13094
- '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.'
13095
- )
13096
- })
13524
+ language_codes: import_zod3.z.array(
13525
+ import_zod3.z.enum([
13526
+ "en",
13527
+ "en_au",
13528
+ "en_uk",
13529
+ "en_us",
13530
+ "es",
13531
+ "fr",
13532
+ "de",
13533
+ "it",
13534
+ "pt",
13535
+ "nl",
13536
+ "af",
13537
+ "sq",
13538
+ "am",
13539
+ "ar",
13540
+ "hy",
13541
+ "as",
13542
+ "az",
13543
+ "ba",
13544
+ "eu",
13545
+ "be",
13546
+ "bn",
13547
+ "bs",
13548
+ "br",
13549
+ "bg",
13550
+ "my",
13551
+ "ca",
13552
+ "zh",
13553
+ "hr",
13554
+ "cs",
13555
+ "da",
13556
+ "et",
13557
+ "fo",
13558
+ "fi",
13559
+ "gl",
13560
+ "ka",
13561
+ "el",
13562
+ "gu",
13563
+ "ht",
13564
+ "ha",
13565
+ "haw",
13566
+ "he",
13567
+ "hi",
13568
+ "hu",
13569
+ "is",
13570
+ "id",
13571
+ "ja",
13572
+ "jw",
13573
+ "kn",
13574
+ "kk",
13575
+ "km",
13576
+ "ko",
13577
+ "lo",
13578
+ "la",
13579
+ "lv",
13580
+ "ln",
13581
+ "lt",
13582
+ "lb",
13583
+ "mk",
13584
+ "mg",
13585
+ "ms",
13586
+ "ml",
13587
+ "mt",
13588
+ "mi",
13589
+ "mr",
13590
+ "mn",
13591
+ "ne",
13592
+ "no",
13593
+ "nn",
13594
+ "oc",
13595
+ "pa",
13596
+ "ps",
13597
+ "fa",
13598
+ "pl",
13599
+ "ro",
13600
+ "ru",
13601
+ "sa",
13602
+ "sr",
13603
+ "sn",
13604
+ "sd",
13605
+ "si",
13606
+ "sk",
13607
+ "sl",
13608
+ "so",
13609
+ "su",
13610
+ "sw",
13611
+ "sv",
13612
+ "tl",
13613
+ "tg",
13614
+ "ta",
13615
+ "tt",
13616
+ "te",
13617
+ "th",
13618
+ "bo",
13619
+ "tr",
13620
+ "tk",
13621
+ "uk",
13622
+ "ur",
13623
+ "uz",
13624
+ "vi",
13625
+ "cy",
13626
+ "yi",
13627
+ "yo"
13628
+ ]).describe(
13629
+ "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"
13630
+ )
13097
13631
  ).nullish().describe(
13098
- "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"
13099
- ),
13100
- confidence: import_zod3.z.number().min(getTranscriptResponseConfidenceMin).max(getTranscriptResponseConfidenceMax).nullish().describe(
13101
- "The confidence score for the transcript, between 0.0 (low confidence) and 1.0 (high confidence)"
13632
+ "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"
13102
13633
  ),
13103
- audio_duration: import_zod3.z.number().nullish().describe("The duration of this transcript object's media file, in seconds"),
13104
- punctuate: import_zod3.z.boolean().nullish().describe("Whether Automatic Punctuation is enabled, either true or false"),
13105
- format_text: import_zod3.z.boolean().nullish().describe("Whether Text Formatting is enabled, either true or false"),
13106
- disfluencies: import_zod3.z.boolean().nullish().describe('Transcribe Filler Words, like "umm", in your media file; can be true or false'),
13107
- multichannel: import_zod3.z.boolean().nullish().describe(
13108
- "Whether [Multichannel transcription](https://www.assemblyai.com/docs/models/speech-recognition#multichannel-transcription) was enabled in the transcription request, either true or false"
13109
- ),
13110
- audio_channels: import_zod3.z.number().optional().describe(
13111
- "The number of audio channels in the audio file. This is only present when multichannel is enabled."
13634
+ language_confidence: import_zod3.z.number().nullable().describe(
13635
+ "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."
13112
13636
  ),
13113
- webhook_url: import_zod3.z.string().nullish().describe(
13114
- "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"
13115
- ),
13116
- webhook_status_code: import_zod3.z.number().nullish().describe(
13117
- "The status code we received from your server when delivering the transcript completed or failed webhook request, if a webhook URL was provided"
13637
+ language_confidence_threshold: import_zod3.z.number().nullable().describe(
13638
+ "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"
13118
13639
  ),
13119
- webhook_auth: import_zod3.z.boolean().describe("Whether webhook authentication details were provided"),
13120
- webhook_auth_header_name: import_zod3.z.string().nullish().describe(
13121
- "The header name to be sent with the transcript completed or failed webhook requests"
13640
+ language_detection: import_zod3.z.boolean().nullish().describe(
13641
+ "Whether [Automatic language detection](/docs/pre-recorded-audio/automatic-language-detection) is enabled, either true or false"
13122
13642
  ),
13123
- speed_boost: import_zod3.z.boolean().nullish().describe("Whether speed boost is enabled"),
13124
- auto_highlights: import_zod3.z.boolean().describe("Whether Key Phrases is enabled, either true or false"),
13125
- auto_highlights_result: import_zod3.z.object({
13126
- status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
13127
- results: import_zod3.z.array(
13128
- import_zod3.z.object({
13129
- count: import_zod3.z.number().describe("The total number of times the key phrase appears in the audio file"),
13130
- rank: import_zod3.z.number().min(getTranscriptResponseAutoHighlightsResultResultsItemRankMin).max(getTranscriptResponseAutoHighlightsResultResultsItemRankMax).describe(
13131
- "The total relevancy to the overall audio file of this key phrase - a greater number means more relevant"
13132
- ),
13133
- text: import_zod3.z.string().describe("The text itself of the key phrase"),
13134
- timestamps: import_zod3.z.array(
13135
- import_zod3.z.object({
13136
- start: import_zod3.z.number().describe("The start time in milliseconds"),
13137
- end: import_zod3.z.number().describe("The end time in milliseconds")
13138
- }).describe("Timestamp containing a start and end property in milliseconds")
13139
- ).describe("The timestamp of the of the key phrase")
13140
- })
13141
- ).describe("A temporally-sequential array of Key Phrases")
13142
- }).describe(
13143
- "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"
13144
- ).or(import_zod3.z.null()).optional().describe(
13145
- "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"
13643
+ language_detection_options: import_zod3.z.object({
13644
+ expected_languages: import_zod3.z.array(import_zod3.z.string()).optional().describe(
13645
+ '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.'
13646
+ ),
13647
+ fallback_language: import_zod3.z.string().default(getTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault).describe(
13648
+ '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'
13649
+ ),
13650
+ code_switching: import_zod3.z.boolean().optional().describe(
13651
+ "Whether [code switching](/docs/speech-to-text/pre-recorded-audio/code-switching) should be detected.\n"
13652
+ ),
13653
+ code_switching_confidence_threshold: import_zod3.z.number().default(
13654
+ getTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault
13655
+ ).describe(
13656
+ "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"
13657
+ )
13658
+ }).optional().describe(
13659
+ "Specify options for [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection)."
13146
13660
  ),
13147
- audio_start_from: import_zod3.z.number().nullish().describe(
13148
- "The point in time, in milliseconds, in the file at which the transcription was started"
13661
+ multichannel: import_zod3.z.boolean().nullish().describe(
13662
+ "Whether [Multichannel transcription](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) was enabled in the transcription request, either true or false"
13149
13663
  ),
13150
- audio_end_at: import_zod3.z.number().nullish().describe(
13151
- "The point in time, in milliseconds, in the file at which the transcription was terminated"
13664
+ prompt: import_zod3.z.string().optional().describe(
13665
+ "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"
13152
13666
  ),
13153
- word_boost: import_zod3.z.array(import_zod3.z.string()).optional().describe("The list of custom vocabulary to boost transcription probability for"),
13154
- boost_param: import_zod3.z.string().nullish().describe("The word boost parameter value"),
13155
- filter_profanity: import_zod3.z.boolean().nullish().describe(
13156
- "Whether [Profanity Filtering](https://www.assemblyai.com/docs/models/speech-recognition#profanity-filtering) is enabled, either true or false"
13667
+ punctuate: import_zod3.z.boolean().nullish().describe(
13668
+ "Whether [Automatic Punctuation](https://www.assemblyai.com/docs/pre-recorded-audio) is enabled, either true or false"
13157
13669
  ),
13158
13670
  redact_pii: import_zod3.z.boolean().describe(
13159
- "Whether [PII Redaction](https://www.assemblyai.com/docs/models/pii-redaction) is enabled, either true or false"
13671
+ "Whether [PII Redaction](https://www.assemblyai.com/docs/pii-redaction) is enabled, either true or false"
13160
13672
  ),
13161
13673
  redact_pii_audio: import_zod3.z.boolean().nullish().describe(
13162
- "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"
13674
+ "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"
13675
+ ),
13676
+ redact_pii_audio_options: import_zod3.z.object({
13677
+ return_redacted_no_speech_audio: import_zod3.z.boolean().optional().describe(
13678
+ "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`."
13679
+ ),
13680
+ override_audio_redaction_method: import_zod3.z.enum(["silence"]).optional().describe(
13681
+ "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."
13682
+ )
13683
+ }).optional().describe(
13684
+ "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"
13163
13685
  ),
13164
13686
  redact_pii_audio_quality: import_zod3.z.enum(["mp3", "wav"]).describe(
13165
- "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."
13687
+ "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."
13166
13688
  ).or(import_zod3.z.null()).optional().describe(
13167
- "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"
13689
+ "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"
13168
13690
  ),
13169
13691
  redact_pii_policies: import_zod3.z.array(
13170
13692
  import_zod3.z.enum([
@@ -13214,19 +13736,307 @@ var getTranscriptResponse = import_zod3.z.object({
13214
13736
  "zodiac_sign"
13215
13737
  ]).describe("The type of PII to redact")
13216
13738
  ).nullish().describe(
13217
- "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"
13739
+ "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"
13218
13740
  ),
13219
13741
  redact_pii_sub: import_zod3.z.enum(["entity_name", "hash"]).optional().describe(
13220
- '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.'
13742
+ "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."
13743
+ ),
13744
+ sentiment_analysis: import_zod3.z.boolean().nullish().describe(
13745
+ "Whether [Sentiment Analysis](https://www.assemblyai.com/docs/speech-understanding/sentiment-analysis) is enabled, can be true or false"
13746
+ ),
13747
+ sentiment_analysis_results: import_zod3.z.array(
13748
+ import_zod3.z.object({
13749
+ text: import_zod3.z.string().describe("The transcript of the sentence"),
13750
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, of the sentence"),
13751
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, of the sentence"),
13752
+ sentiment: import_zod3.z.enum(["POSITIVE", "NEUTRAL", "NEGATIVE"]),
13753
+ confidence: import_zod3.z.number().describe(
13754
+ "The confidence score for the detected sentiment of the sentence, from 0 to 1"
13755
+ ),
13756
+ channel: import_zod3.z.string().nullish().describe(
13757
+ "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13758
+ ),
13759
+ speaker: import_zod3.z.string().nullable().describe(
13760
+ "The speaker of the sentence if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
13761
+ )
13762
+ }).describe("The result of the Sentiment Analysis model")
13763
+ ).nullish().describe(
13764
+ "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"
13221
13765
  ),
13222
13766
  speaker_labels: import_zod3.z.boolean().nullish().describe(
13223
- "Whether [Speaker diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, can be true or false"
13767
+ "Whether [Speaker diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, can be true or false"
13224
13768
  ),
13225
13769
  speakers_expected: import_zod3.z.number().nullish().describe(
13226
- "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."
13770
+ "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."
13771
+ ),
13772
+ speech_model_used: import_zod3.z.string().optional().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
+ speech_models: import_zod3.z.array(
13776
+ import_zod3.z.string().describe(
13777
+ "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."
13778
+ )
13779
+ ).nullish().describe(
13780
+ "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"
13781
+ ),
13782
+ speech_threshold: import_zod3.z.number().nullish().describe(
13783
+ "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"
13784
+ ),
13785
+ speech_understanding: import_zod3.z.object({
13786
+ request: import_zod3.z.object({
13787
+ translation: import_zod3.z.object({
13788
+ target_languages: import_zod3.z.array(import_zod3.z.string()).describe(
13789
+ 'List of target language codes (e.g., `["es", "de"]`). See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for supported languages.'
13790
+ ),
13791
+ formal: import_zod3.z.boolean().default(getTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault).describe(
13792
+ "Use formal language style. See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for more details."
13793
+ ),
13794
+ match_original_utterance: import_zod3.z.boolean().optional().describe(
13795
+ "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."
13796
+ )
13797
+ })
13798
+ }).describe(
13799
+ "Request body for [Translation](https://www.assemblyai.com/docs/speech-understanding/translation)."
13800
+ ).or(
13801
+ import_zod3.z.object({
13802
+ speaker_identification: import_zod3.z.object({
13803
+ speaker_type: import_zod3.z.enum(["role", "name"]).describe(
13804
+ "Type of speaker identification. See [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification) for details on each type."
13805
+ ),
13806
+ known_values: import_zod3.z.array(import_zod3.z.string()).optional().describe(
13807
+ 'Required if speaker_type is "role". Each value must be 35 characters or less.'
13808
+ ),
13809
+ speakers: import_zod3.z.array(
13810
+ import_zod3.z.object({
13811
+ role: import_zod3.z.string().optional().describe(
13812
+ 'The role of the speaker. Required when `speaker_type` is "role".'
13813
+ ),
13814
+ name: import_zod3.z.string().optional().describe(
13815
+ 'The name of the speaker. Required when `speaker_type` is "name".'
13816
+ ),
13817
+ description: import_zod3.z.string().optional().describe(
13818
+ "A description of the speaker to help the model identify them based on conversational context."
13819
+ )
13820
+ })
13821
+ ).optional().describe(
13822
+ "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."
13823
+ )
13824
+ })
13825
+ }).describe(
13826
+ "Request body for [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification)."
13827
+ )
13828
+ ).or(
13829
+ import_zod3.z.object({
13830
+ custom_formatting: import_zod3.z.object({
13831
+ date: import_zod3.z.string().optional().describe(
13832
+ 'Date format pattern (e.g., `"mm/dd/yyyy"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
13833
+ ),
13834
+ phone_number: import_zod3.z.string().optional().describe(
13835
+ 'Phone number format pattern (e.g., `"(xxx)xxx-xxxx"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
13836
+ ),
13837
+ email: import_zod3.z.string().optional().describe(
13838
+ 'Email format pattern (e.g., `"username@domain.com"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
13839
+ )
13840
+ })
13841
+ }).describe(
13842
+ "Request body for [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting)."
13843
+ )
13844
+ ).optional(),
13845
+ response: import_zod3.z.object({
13846
+ translation: import_zod3.z.object({
13847
+ status: import_zod3.z.string().optional()
13848
+ }).optional()
13849
+ }).or(
13850
+ import_zod3.z.object({
13851
+ speaker_identification: import_zod3.z.object({
13852
+ mapping: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional().describe(
13853
+ 'A mapping of the original generic speaker labels (e.g., "A", "B") to the identified speaker names or roles.'
13854
+ ),
13855
+ status: import_zod3.z.string().optional()
13856
+ }).optional()
13857
+ })
13858
+ ).or(
13859
+ import_zod3.z.object({
13860
+ custom_formatting: import_zod3.z.object({
13861
+ mapping: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional(),
13862
+ formatted_text: import_zod3.z.string().optional()
13863
+ }).optional()
13864
+ })
13865
+ ).optional()
13866
+ }).optional().describe(
13867
+ "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"
13868
+ ),
13869
+ status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).describe(
13870
+ "The status of your transcript. Possible values are queued, processing, completed, or error."
13871
+ ),
13872
+ summarization: import_zod3.z.boolean().describe(
13873
+ "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"
13874
+ ),
13875
+ summary: import_zod3.z.string().nullish().describe(
13876
+ "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."
13877
+ ),
13878
+ summary_model: import_zod3.z.string().nullish().describe(
13879
+ "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"
13880
+ ),
13881
+ summary_type: import_zod3.z.string().nullish().describe(
13882
+ "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."
13883
+ ),
13884
+ remove_audio_tags: import_zod3.z.enum(["all"]).describe(
13885
+ "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"
13886
+ ).or(import_zod3.z.null()).optional().describe(
13887
+ "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"
13888
+ ),
13889
+ temperature: import_zod3.z.number().nullish().describe(
13890
+ "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"
13891
+ ),
13892
+ text: import_zod3.z.string().nullish().describe("The textual transcript of your media file"),
13893
+ throttled: import_zod3.z.boolean().nullish().describe(
13894
+ "True while a request is throttled and false when a request is no longer throttled"
13895
+ ),
13896
+ utterances: import_zod3.z.array(
13897
+ import_zod3.z.object({
13898
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this utterance"),
13899
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, of the utterance in the audio file"),
13900
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, of the utterance in the audio file"),
13901
+ text: import_zod3.z.string().describe("The text for this utterance"),
13902
+ words: import_zod3.z.array(
13903
+ import_zod3.z.object({
13904
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
13905
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
13906
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
13907
+ text: import_zod3.z.string().describe("The text of the word"),
13908
+ channel: import_zod3.z.string().nullish().describe(
13909
+ "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13910
+ ),
13911
+ speaker: import_zod3.z.string().nullable().describe(
13912
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
13913
+ )
13914
+ })
13915
+ ).describe("The words in the utterance."),
13916
+ channel: import_zod3.z.string().nullish().describe(
13917
+ "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13918
+ ),
13919
+ speaker: import_zod3.z.string().describe(
13920
+ '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.'
13921
+ ),
13922
+ translated_texts: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional().describe(
13923
+ 'Translations keyed by language code (e.g., `{"es": "Texto traducido", "de": "\xDCbersetzter Text"}`). Only present when `match_original_utterance` is enabled with translation.'
13924
+ )
13925
+ })
13926
+ ).nullish().describe(
13927
+ "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"
13928
+ ),
13929
+ webhook_auth: import_zod3.z.boolean().describe(
13930
+ "Whether [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) authentication details were provided"
13931
+ ),
13932
+ webhook_auth_header_name: import_zod3.z.string().nullish().describe(
13933
+ "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"
13934
+ ),
13935
+ webhook_status_code: import_zod3.z.number().nullish().describe(
13936
+ "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"
13937
+ ),
13938
+ webhook_url: import_zod3.z.string().nullish().describe(
13939
+ "The URL to which we send [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) requests.\n"
13940
+ ),
13941
+ words: import_zod3.z.array(
13942
+ import_zod3.z.object({
13943
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
13944
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
13945
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
13946
+ text: import_zod3.z.string().describe("The text of the word"),
13947
+ channel: import_zod3.z.string().nullish().describe(
13948
+ "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13949
+ ),
13950
+ speaker: import_zod3.z.string().nullable().describe(
13951
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
13952
+ )
13953
+ })
13954
+ ).nullish().describe(
13955
+ "An array of temporally-sequential word objects, one for each word in the transcript.\n"
13956
+ ),
13957
+ acoustic_model: import_zod3.z.string().describe("This parameter does not currently have any functionality attached to it."),
13958
+ custom_topics: import_zod3.z.boolean().nullish().describe("This parameter does not currently have any functionality attached to it."),
13959
+ language_model: import_zod3.z.string().describe("This parameter does not currently have any functionality attached to it."),
13960
+ speech_model: import_zod3.z.string().describe(
13961
+ "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."
13962
+ ).or(import_zod3.z.null()).describe(
13963
+ "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"
13964
+ ),
13965
+ speed_boost: import_zod3.z.boolean().nullish().describe("This parameter does not currently have any functionality attached to it."),
13966
+ topics: import_zod3.z.array(import_zod3.z.string()).optional().describe("This parameter does not currently have any functionality attached to it."),
13967
+ translated_texts: import_zod3.z.object({
13968
+ language_code: import_zod3.z.string().optional().describe("Translated text for this language code")
13969
+ }).optional().describe(
13970
+ "Translated text keyed by language code. See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for more details."
13971
+ )
13972
+ }).describe("A transcript object");
13973
+ var deleteTranscriptParams = import_zod3.z.object({
13974
+ transcript_id: import_zod3.z.string().describe("ID of the transcript")
13975
+ });
13976
+ var deleteTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault = "auto";
13977
+ var deleteTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault = false;
13978
+ var deleteTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault = 0.3;
13979
+ var deleteTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault = false;
13980
+ var deleteTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault = true;
13981
+ var deleteTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault = false;
13982
+ var deleteTranscriptResponse = import_zod3.z.object({
13983
+ audio_channels: import_zod3.z.number().optional().describe(
13984
+ "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."
13985
+ ),
13986
+ audio_duration: import_zod3.z.number().nullish().describe("The duration of this transcript object's media file, in seconds"),
13987
+ audio_end_at: import_zod3.z.number().nullish().describe(
13988
+ "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."
13989
+ ),
13990
+ audio_start_from: import_zod3.z.number().nullish().describe(
13991
+ "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."
13992
+ ),
13993
+ audio_url: import_zod3.z.string().describe("The URL of the media that was transcribed"),
13994
+ auto_chapters: import_zod3.z.boolean().nullish().describe(
13995
+ "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"
13996
+ ),
13997
+ auto_highlights: import_zod3.z.boolean().describe(
13998
+ "Whether [Key Phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases) is enabled, either true or false"
13999
+ ),
14000
+ auto_highlights_result: import_zod3.z.object({
14001
+ status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
14002
+ results: import_zod3.z.array(
14003
+ import_zod3.z.object({
14004
+ count: import_zod3.z.number().describe("The total number of times the key phrase appears in the audio file"),
14005
+ rank: import_zod3.z.number().describe(
14006
+ "The total relevancy to the overall audio file of this key phrase - a greater number means more relevant"
14007
+ ),
14008
+ text: import_zod3.z.string().describe("The text itself of the key phrase"),
14009
+ timestamps: import_zod3.z.array(
14010
+ import_zod3.z.object({
14011
+ start: import_zod3.z.number().describe("The start time in milliseconds"),
14012
+ end: import_zod3.z.number().describe("The end time in milliseconds")
14013
+ }).describe("Timestamp containing a start and end property in milliseconds")
14014
+ ).describe("The timestamp of the of the key phrase")
14015
+ })
14016
+ ).describe("A temporally-sequential array of Key Phrases")
14017
+ }).describe(
14018
+ "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"
14019
+ ).or(import_zod3.z.null()).optional().describe(
14020
+ "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"
14021
+ ),
14022
+ chapters: import_zod3.z.array(
14023
+ import_zod3.z.object({
14024
+ gist: import_zod3.z.string().describe(
14025
+ "An ultra-short summary (just a few words) of the content spoken in the chapter"
14026
+ ),
14027
+ headline: import_zod3.z.string().describe("A single sentence summary of the content spoken during the chapter"),
14028
+ summary: import_zod3.z.string().describe("A one paragraph summary of the content spoken during the chapter"),
14029
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter"),
14030
+ end: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter")
14031
+ }).describe("Chapter of the audio file")
14032
+ ).nullish().describe(
14033
+ "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."
14034
+ ),
14035
+ confidence: import_zod3.z.number().nullish().describe(
14036
+ "The confidence score for the transcript, between 0.0 (low confidence) and 1.0 (high confidence)"
13227
14037
  ),
13228
14038
  content_safety: import_zod3.z.boolean().nullish().describe(
13229
- "Whether [Content Moderation](https://www.assemblyai.com/docs/models/content-moderation) is enabled, can be true or false"
14039
+ "Whether [Content Moderation](https://www.assemblyai.com/docs/content-moderation) is enabled, can be true or false"
13230
14040
  ),
13231
14041
  content_safety_labels: import_zod3.z.object({
13232
14042
  status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
@@ -13236,12 +14046,8 @@ var getTranscriptResponse = import_zod3.z.object({
13236
14046
  labels: import_zod3.z.array(
13237
14047
  import_zod3.z.object({
13238
14048
  label: import_zod3.z.string().describe("The label of the sensitive topic"),
13239
- confidence: import_zod3.z.number().min(
13240
- getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin
13241
- ).max(
13242
- getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax
13243
- ).describe("The confidence score for the topic being discussed, from 0 to 1"),
13244
- severity: import_zod3.z.number().min(getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin).max(getTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax).describe("How severely the topic is discussed in the section, from 0 to 1")
14049
+ confidence: import_zod3.z.number().describe("The confidence score for the topic being discussed, from 0 to 1"),
14050
+ severity: import_zod3.z.number().describe("How severely the topic is discussed in the section, from 0 to 1")
13245
14051
  })
13246
14052
  ).describe(
13247
14053
  "An array of safety labels, one per sensitive topic that was detected in the section"
@@ -13254,128 +14060,39 @@ var getTranscriptResponse = import_zod3.z.object({
13254
14060
  }).describe("Timestamp containing a start and end property in milliseconds")
13255
14061
  })
13256
14062
  ).describe("An array of results for the Content Moderation model"),
13257
- summary: import_zod3.z.record(
13258
- import_zod3.z.string(),
13259
- import_zod3.z.number().min(getTranscriptResponseContentSafetyLabelsSummaryMinOne).max(getTranscriptResponseContentSafetyLabelsSummaryMaxOne).describe(
13260
- 'A confidence score for the presence of the sensitive topic "topic" across the entire audio file'
13261
- )
13262
- ).describe(
14063
+ summary: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.number()).describe(
13263
14064
  "A summary of the Content Moderation confidence results for the entire audio file"
13264
14065
  ),
13265
14066
  severity_score_summary: import_zod3.z.record(
13266
14067
  import_zod3.z.string(),
13267
14068
  import_zod3.z.object({
13268
- low: import_zod3.z.number().min(getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin).max(getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax),
13269
- medium: import_zod3.z.number().min(getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin).max(getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax),
13270
- high: import_zod3.z.number().min(getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin).max(getTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax)
14069
+ low: import_zod3.z.number(),
14070
+ medium: import_zod3.z.number(),
14071
+ high: import_zod3.z.number()
13271
14072
  })
13272
14073
  ).describe(
13273
14074
  "A summary of the Content Moderation severity results for the entire audio file"
13274
14075
  )
13275
14076
  }).describe(
13276
- "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"
13277
- ).or(import_zod3.z.null()).optional().describe(
13278
- "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"
13279
- ),
13280
- iab_categories: import_zod3.z.boolean().nullish().describe(
13281
- "Whether [Topic Detection](https://www.assemblyai.com/docs/models/topic-detection) is enabled, can be true or false"
13282
- ),
13283
- iab_categories_result: import_zod3.z.object({
13284
- status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
13285
- results: import_zod3.z.array(
13286
- import_zod3.z.object({
13287
- text: import_zod3.z.string().describe("The text in the transcript in which a detected topic occurs"),
13288
- labels: import_zod3.z.array(
13289
- import_zod3.z.object({
13290
- relevance: import_zod3.z.number().min(
13291
- getTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin
13292
- ).max(
13293
- getTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax
13294
- ).describe("How relevant the detected topic is of a detected topic"),
13295
- label: import_zod3.z.string().describe(
13296
- "The IAB taxonomical label for the label of the detected topic, where > denotes supertopic/subtopic relationship"
13297
- )
13298
- })
13299
- ).optional().describe("An array of detected topics in the text"),
13300
- timestamp: import_zod3.z.object({
13301
- start: import_zod3.z.number().describe("The start time in milliseconds"),
13302
- end: import_zod3.z.number().describe("The end time in milliseconds")
13303
- }).optional().describe("Timestamp containing a start and end property in milliseconds")
13304
- }).describe("The result of the topic detection model")
13305
- ).describe("An array of results for the Topic Detection model"),
13306
- summary: import_zod3.z.record(
13307
- import_zod3.z.string(),
13308
- import_zod3.z.number().min(getTranscriptResponseIabCategoriesResultSummaryMinOne).max(getTranscriptResponseIabCategoriesResultSummaryMaxOne)
13309
- ).describe("The overall relevance of topic to the entire audio file")
13310
- }).describe(
13311
- "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"
14077
+ "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"
13312
14078
  ).or(import_zod3.z.null()).optional().describe(
13313
- "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"
14079
+ "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"
13314
14080
  ),
13315
14081
  custom_spelling: import_zod3.z.array(
13316
14082
  import_zod3.z.object({
13317
- from: import_zod3.z.array(import_zod3.z.string().describe("Word or phrase to replace")).describe("Words or phrases to replace"),
14083
+ from: import_zod3.z.array(import_zod3.z.string()).describe("Words or phrases to replace"),
13318
14084
  to: import_zod3.z.string().describe("Word to replace with")
13319
14085
  }).describe(
13320
14086
  "Object containing words or phrases to replace, and the word or phrase to replace with"
13321
14087
  )
13322
- ).nullish().describe("Customize how words are spelled and formatted using to and from values"),
13323
- keyterms_prompt: import_zod3.z.array(import_zod3.z.string()).optional().describe(
13324
- "Improve accuracy with up to 1000 domain-specific words or phrases (maximum 6 words per phrase).\n"
13325
- ),
13326
- prompt: import_zod3.z.string().optional().describe("This parameter does not currently have any functionality attached to it."),
13327
- auto_chapters: import_zod3.z.boolean().nullish().describe(
13328
- "Whether [Auto Chapters](https://www.assemblyai.com/docs/models/auto-chapters) is enabled, can be true or false"
13329
- ),
13330
- chapters: import_zod3.z.array(
13331
- import_zod3.z.object({
13332
- gist: import_zod3.z.string().describe(
13333
- "An ultra-short summary (just a few words) of the content spoken in the chapter"
13334
- ),
13335
- headline: import_zod3.z.string().describe("A single sentence summary of the content spoken during the chapter"),
13336
- summary: import_zod3.z.string().describe("A one paragraph summary of the content spoken during the chapter"),
13337
- start: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter"),
13338
- end: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter")
13339
- }).describe("Chapter of the audio file")
13340
- ).nullish().describe("An array of temporally sequential chapters for the audio file"),
13341
- summarization: import_zod3.z.boolean().describe(
13342
- "Whether [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled, either true or false"
13343
- ),
13344
- summary_type: import_zod3.z.string().nullish().describe(
13345
- "The type of summary generated, if [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled"
13346
- ),
13347
- summary_model: import_zod3.z.string().nullish().describe(
13348
- "The Summarization model used to generate the summary,\nif [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled\n"
13349
- ),
13350
- summary: import_zod3.z.string().nullish().describe(
13351
- "The generated summary of the media file, if [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled"
13352
- ),
13353
- custom_topics: import_zod3.z.boolean().nullish().describe("Whether custom topics is enabled, either true or false"),
13354
- topics: import_zod3.z.array(import_zod3.z.string()).optional().describe("The list of custom topics provided if custom topics is enabled"),
13355
- sentiment_analysis: import_zod3.z.boolean().nullish().describe(
13356
- "Whether [Sentiment Analysis](https://www.assemblyai.com/docs/models/sentiment-analysis) is enabled, can be true or false"
13357
- ),
13358
- sentiment_analysis_results: import_zod3.z.array(
13359
- import_zod3.z.object({
13360
- text: import_zod3.z.string().describe("The transcript of the sentence"),
13361
- start: import_zod3.z.number().describe("The starting time, in milliseconds, of the sentence"),
13362
- end: import_zod3.z.number().describe("The ending time, in milliseconds, of the sentence"),
13363
- sentiment: import_zod3.z.enum(["POSITIVE", "NEUTRAL", "NEGATIVE"]),
13364
- confidence: import_zod3.z.number().min(getTranscriptResponseSentimentAnalysisResultsItemConfidenceMin).max(getTranscriptResponseSentimentAnalysisResultsItemConfidenceMax).describe(
13365
- "The confidence score for the detected sentiment of the sentence, from 0 to 1"
13366
- ),
13367
- channel: import_zod3.z.string().nullish().describe(
13368
- "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13369
- ),
13370
- speaker: import_zod3.z.string().nullable().describe(
13371
- "The speaker of the sentence if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
13372
- )
13373
- }).describe("The result of the Sentiment Analysis model")
13374
14088
  ).nullish().describe(
13375
- "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"
14089
+ "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."
13376
14090
  ),
13377
- entity_detection: import_zod3.z.boolean().nullish().describe(
13378
- "Whether [Entity Detection](https://www.assemblyai.com/docs/models/entity-detection) is enabled, can be true or false"
14091
+ disfluencies: import_zod3.z.boolean().nullish().describe(
14092
+ 'Transcribe [Filler Words](https://www.assemblyai.com/docs/pre-recorded-audio/filler-words), like "umm", in your media file; can be true or false'
14093
+ ),
14094
+ domain: import_zod3.z.string().nullish().describe(
14095
+ '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'
13379
14096
  ),
13380
14097
  entities: import_zod3.z.array(
13381
14098
  import_zod3.z.object({
@@ -13434,61 +14151,49 @@ var getTranscriptResponse = import_zod3.z.object({
13434
14151
  )
13435
14152
  }).describe("A detected entity")
13436
14153
  ).nullish().describe(
13437
- "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"
13438
- ),
13439
- speech_threshold: import_zod3.z.number().min(getTranscriptResponseSpeechThresholdMin).max(getTranscriptResponseSpeechThresholdMax).nullish().describe(
13440
- "Defaults to null. Reject audio files that contain less than this fraction of speech.\nValid values are in the range [0, 1] inclusive.\n"
14154
+ "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"
13441
14155
  ),
13442
- throttled: import_zod3.z.boolean().nullish().describe(
13443
- "True while a request is throttled and false when a request is no longer throttled"
14156
+ entity_detection: import_zod3.z.boolean().nullish().describe(
14157
+ "Whether [Entity Detection](https://www.assemblyai.com/docs/speech-understanding/entity-detection) is enabled, can be true or false"
13444
14158
  ),
13445
14159
  error: import_zod3.z.string().optional().describe("Error message of why the transcript failed"),
13446
- language_model: import_zod3.z.string().describe("The language model that was used for the transcript"),
13447
- acoustic_model: import_zod3.z.string().describe("The acoustic model that was used for the transcript")
13448
- }).describe("A transcript object");
13449
- var deleteTranscriptParams = import_zod3.z.object({
13450
- transcript_id: import_zod3.z.string().describe("ID of the transcript")
13451
- });
13452
- var deleteTranscriptResponseLanguageConfidenceThresholdMin = 0;
13453
- var deleteTranscriptResponseLanguageConfidenceThresholdMax = 1;
13454
- var deleteTranscriptResponseLanguageConfidenceMin = 0;
13455
- var deleteTranscriptResponseLanguageConfidenceMax = 1;
13456
- var deleteTranscriptResponseSpeechModelDefault = null;
13457
- var deleteTranscriptResponseWordsItemConfidenceMin = 0;
13458
- var deleteTranscriptResponseWordsItemConfidenceMax = 1;
13459
- var deleteTranscriptResponseUtterancesItemConfidenceMin = 0;
13460
- var deleteTranscriptResponseUtterancesItemConfidenceMax = 1;
13461
- var deleteTranscriptResponseUtterancesItemWordsItemConfidenceMin = 0;
13462
- var deleteTranscriptResponseUtterancesItemWordsItemConfidenceMax = 1;
13463
- var deleteTranscriptResponseConfidenceMin = 0;
13464
- var deleteTranscriptResponseConfidenceMax = 1;
13465
- var deleteTranscriptResponseAutoHighlightsResultResultsItemRankMin = 0;
13466
- var deleteTranscriptResponseAutoHighlightsResultResultsItemRankMax = 1;
13467
- var deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin = 0;
13468
- var deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax = 1;
13469
- var deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin = 0;
13470
- var deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax = 1;
13471
- var deleteTranscriptResponseContentSafetyLabelsSummaryMinOne = 0;
13472
- var deleteTranscriptResponseContentSafetyLabelsSummaryMaxOne = 1;
13473
- var deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin = 0;
13474
- var deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax = 1;
13475
- var deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin = 0;
13476
- var deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax = 1;
13477
- var deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin = 0;
13478
- var deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax = 1;
13479
- var deleteTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin = 0;
13480
- var deleteTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax = 1;
13481
- var deleteTranscriptResponseIabCategoriesResultSummaryMinOne = 0;
13482
- var deleteTranscriptResponseIabCategoriesResultSummaryMaxOne = 1;
13483
- var deleteTranscriptResponseSentimentAnalysisResultsItemConfidenceMin = 0;
13484
- var deleteTranscriptResponseSentimentAnalysisResultsItemConfidenceMax = 1;
13485
- var deleteTranscriptResponseSpeechThresholdMin = 0;
13486
- var deleteTranscriptResponseSpeechThresholdMax = 1;
13487
- var deleteTranscriptResponse = import_zod3.z.object({
14160
+ filter_profanity: import_zod3.z.boolean().nullish().describe(
14161
+ "Whether [Profanity Filtering](https://www.assemblyai.com/docs/profanity-filtering) is enabled, either true or false"
14162
+ ),
14163
+ format_text: import_zod3.z.boolean().nullish().describe(
14164
+ "Whether [Text Formatting](https://www.assemblyai.com/docs/pre-recorded-audio) is enabled, either true or false"
14165
+ ),
14166
+ iab_categories: import_zod3.z.boolean().nullish().describe(
14167
+ "Whether [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection) is enabled, can be true or false"
14168
+ ),
14169
+ iab_categories_result: import_zod3.z.object({
14170
+ status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
14171
+ results: import_zod3.z.array(
14172
+ import_zod3.z.object({
14173
+ text: import_zod3.z.string().describe("The text in the transcript in which a detected topic occurs"),
14174
+ labels: import_zod3.z.array(
14175
+ import_zod3.z.object({
14176
+ relevance: import_zod3.z.number().describe("How relevant the detected topic is of a detected topic"),
14177
+ label: import_zod3.z.string().describe(
14178
+ "The IAB taxonomical label for the label of the detected topic, where > denotes supertopic/subtopic relationship"
14179
+ )
14180
+ })
14181
+ ).optional().describe("An array of detected topics in the text"),
14182
+ timestamp: import_zod3.z.object({
14183
+ start: import_zod3.z.number().describe("The start time in milliseconds"),
14184
+ end: import_zod3.z.number().describe("The end time in milliseconds")
14185
+ }).optional().describe("Timestamp containing a start and end property in milliseconds")
14186
+ }).describe("The result of the topic detection model")
14187
+ ).describe("An array of results for the Topic Detection model"),
14188
+ summary: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.number()).describe("The overall relevance of topic to the entire audio file")
14189
+ }).describe(
14190
+ "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"
14191
+ ).or(import_zod3.z.null()).optional().describe(
14192
+ "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"
14193
+ ),
13488
14194
  id: import_zod3.z.string().uuid().describe("The unique identifier of your transcript"),
13489
- audio_url: import_zod3.z.string().describe("The URL of the media that was transcribed"),
13490
- status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).describe(
13491
- "The status of your transcript. Possible values are queued, processing, completed, or error."
14195
+ keyterms_prompt: import_zod3.z.array(import_zod3.z.string()).optional().describe(
14196
+ "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"
13492
14197
  ),
13493
14198
  language_code: import_zod3.z.enum([
13494
14199
  "en",
@@ -13593,138 +14298,175 @@ var deleteTranscriptResponse = import_zod3.z.object({
13593
14298
  "cy",
13594
14299
  "yi",
13595
14300
  "yo"
13596
- ]).describe(
13597
- "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"
13598
- ).or(import_zod3.z.string()).optional().describe(
13599
- "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"
13600
- ),
13601
- language_detection: import_zod3.z.boolean().nullish().describe(
13602
- "Whether [Automatic language detection](https://www.assemblyai.com/docs/models/speech-recognition#automatic-language-detection) is enabled, either true or false"
13603
- ),
13604
- language_confidence_threshold: import_zod3.z.number().min(deleteTranscriptResponseLanguageConfidenceThresholdMin).max(deleteTranscriptResponseLanguageConfidenceThresholdMax).nullable().describe(
13605
- "The confidence threshold for the automatically detected language.\nAn error will be returned if the language confidence is below this threshold.\n"
13606
- ),
13607
- language_confidence: import_zod3.z.number().min(deleteTranscriptResponseLanguageConfidenceMin).max(deleteTranscriptResponseLanguageConfidenceMax).nullable().describe(
13608
- "The confidence score for the detected language, between 0.0 (low confidence) and 1.0 (high confidence)"
13609
- ),
13610
- 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(
13611
- "The speech model used for the transcription. When `null`, the default model is used."
13612
- ),
13613
- text: import_zod3.z.string().nullish().describe("The textual transcript of your media file"),
13614
- words: import_zod3.z.array(
13615
- import_zod3.z.object({
13616
- confidence: import_zod3.z.number().min(deleteTranscriptResponseWordsItemConfidenceMin).max(deleteTranscriptResponseWordsItemConfidenceMax).describe("The confidence score for the transcript of this word"),
13617
- start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
13618
- end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
13619
- text: import_zod3.z.string().describe("The text of the word"),
13620
- channel: import_zod3.z.string().nullish().describe(
13621
- "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13622
- ),
13623
- speaker: import_zod3.z.string().nullable().describe(
13624
- "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
13625
- )
13626
- })
13627
- ).nullish().describe(
13628
- "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"
14301
+ ]).optional().describe(
14302
+ "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"
13629
14303
  ),
13630
- utterances: import_zod3.z.array(
13631
- import_zod3.z.object({
13632
- confidence: import_zod3.z.number().min(deleteTranscriptResponseUtterancesItemConfidenceMin).max(deleteTranscriptResponseUtterancesItemConfidenceMax).describe("The confidence score for the transcript of this utterance"),
13633
- start: import_zod3.z.number().describe("The starting time, in milliseconds, of the utterance in the audio file"),
13634
- end: import_zod3.z.number().describe("The ending time, in milliseconds, of the utterance in the audio file"),
13635
- text: import_zod3.z.string().describe("The text for this utterance"),
13636
- words: import_zod3.z.array(
13637
- import_zod3.z.object({
13638
- confidence: import_zod3.z.number().min(deleteTranscriptResponseUtterancesItemWordsItemConfidenceMin).max(deleteTranscriptResponseUtterancesItemWordsItemConfidenceMax).describe("The confidence score for the transcript of this word"),
13639
- start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
13640
- end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
13641
- text: import_zod3.z.string().describe("The text of the word"),
13642
- channel: import_zod3.z.string().nullish().describe(
13643
- "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13644
- ),
13645
- speaker: import_zod3.z.string().nullable().describe(
13646
- "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
13647
- )
13648
- })
13649
- ).describe("The words in the utterance."),
13650
- channel: import_zod3.z.string().nullish().describe(
13651
- "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13652
- ),
13653
- speaker: import_zod3.z.string().describe(
13654
- '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.'
13655
- )
13656
- })
14304
+ language_codes: import_zod3.z.array(
14305
+ import_zod3.z.enum([
14306
+ "en",
14307
+ "en_au",
14308
+ "en_uk",
14309
+ "en_us",
14310
+ "es",
14311
+ "fr",
14312
+ "de",
14313
+ "it",
14314
+ "pt",
14315
+ "nl",
14316
+ "af",
14317
+ "sq",
14318
+ "am",
14319
+ "ar",
14320
+ "hy",
14321
+ "as",
14322
+ "az",
14323
+ "ba",
14324
+ "eu",
14325
+ "be",
14326
+ "bn",
14327
+ "bs",
14328
+ "br",
14329
+ "bg",
14330
+ "my",
14331
+ "ca",
14332
+ "zh",
14333
+ "hr",
14334
+ "cs",
14335
+ "da",
14336
+ "et",
14337
+ "fo",
14338
+ "fi",
14339
+ "gl",
14340
+ "ka",
14341
+ "el",
14342
+ "gu",
14343
+ "ht",
14344
+ "ha",
14345
+ "haw",
14346
+ "he",
14347
+ "hi",
14348
+ "hu",
14349
+ "is",
14350
+ "id",
14351
+ "ja",
14352
+ "jw",
14353
+ "kn",
14354
+ "kk",
14355
+ "km",
14356
+ "ko",
14357
+ "lo",
14358
+ "la",
14359
+ "lv",
14360
+ "ln",
14361
+ "lt",
14362
+ "lb",
14363
+ "mk",
14364
+ "mg",
14365
+ "ms",
14366
+ "ml",
14367
+ "mt",
14368
+ "mi",
14369
+ "mr",
14370
+ "mn",
14371
+ "ne",
14372
+ "no",
14373
+ "nn",
14374
+ "oc",
14375
+ "pa",
14376
+ "ps",
14377
+ "fa",
14378
+ "pl",
14379
+ "ro",
14380
+ "ru",
14381
+ "sa",
14382
+ "sr",
14383
+ "sn",
14384
+ "sd",
14385
+ "si",
14386
+ "sk",
14387
+ "sl",
14388
+ "so",
14389
+ "su",
14390
+ "sw",
14391
+ "sv",
14392
+ "tl",
14393
+ "tg",
14394
+ "ta",
14395
+ "tt",
14396
+ "te",
14397
+ "th",
14398
+ "bo",
14399
+ "tr",
14400
+ "tk",
14401
+ "uk",
14402
+ "ur",
14403
+ "uz",
14404
+ "vi",
14405
+ "cy",
14406
+ "yi",
14407
+ "yo"
14408
+ ]).describe(
14409
+ "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"
14410
+ )
13657
14411
  ).nullish().describe(
13658
- "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"
13659
- ),
13660
- confidence: import_zod3.z.number().min(deleteTranscriptResponseConfidenceMin).max(deleteTranscriptResponseConfidenceMax).nullish().describe(
13661
- "The confidence score for the transcript, between 0.0 (low confidence) and 1.0 (high confidence)"
13662
- ),
13663
- audio_duration: import_zod3.z.number().nullish().describe("The duration of this transcript object's media file, in seconds"),
13664
- punctuate: import_zod3.z.boolean().nullish().describe("Whether Automatic Punctuation is enabled, either true or false"),
13665
- format_text: import_zod3.z.boolean().nullish().describe("Whether Text Formatting is enabled, either true or false"),
13666
- disfluencies: import_zod3.z.boolean().nullish().describe('Transcribe Filler Words, like "umm", in your media file; can be true or false'),
13667
- multichannel: import_zod3.z.boolean().nullish().describe(
13668
- "Whether [Multichannel transcription](https://www.assemblyai.com/docs/models/speech-recognition#multichannel-transcription) was enabled in the transcription request, either true or false"
13669
- ),
13670
- audio_channels: import_zod3.z.number().optional().describe(
13671
- "The number of audio channels in the audio file. This is only present when multichannel is enabled."
14412
+ "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"
13672
14413
  ),
13673
- webhook_url: import_zod3.z.string().nullish().describe(
13674
- "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"
14414
+ language_confidence: import_zod3.z.number().nullable().describe(
14415
+ "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."
13675
14416
  ),
13676
- webhook_status_code: import_zod3.z.number().nullish().describe(
13677
- "The status code we received from your server when delivering the transcript completed or failed webhook request, if a webhook URL was provided"
14417
+ language_confidence_threshold: import_zod3.z.number().nullable().describe(
14418
+ "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"
13678
14419
  ),
13679
- webhook_auth: import_zod3.z.boolean().describe("Whether webhook authentication details were provided"),
13680
- webhook_auth_header_name: import_zod3.z.string().nullish().describe(
13681
- "The header name to be sent with the transcript completed or failed webhook requests"
14420
+ language_detection: import_zod3.z.boolean().nullish().describe(
14421
+ "Whether [Automatic language detection](/docs/pre-recorded-audio/automatic-language-detection) is enabled, either true or false"
13682
14422
  ),
13683
- speed_boost: import_zod3.z.boolean().nullish().describe("Whether speed boost is enabled"),
13684
- auto_highlights: import_zod3.z.boolean().describe("Whether Key Phrases is enabled, either true or false"),
13685
- auto_highlights_result: import_zod3.z.object({
13686
- status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
13687
- results: import_zod3.z.array(
13688
- import_zod3.z.object({
13689
- count: import_zod3.z.number().describe("The total number of times the key phrase appears in the audio file"),
13690
- rank: import_zod3.z.number().min(deleteTranscriptResponseAutoHighlightsResultResultsItemRankMin).max(deleteTranscriptResponseAutoHighlightsResultResultsItemRankMax).describe(
13691
- "The total relevancy to the overall audio file of this key phrase - a greater number means more relevant"
13692
- ),
13693
- text: import_zod3.z.string().describe("The text itself of the key phrase"),
13694
- timestamps: import_zod3.z.array(
13695
- import_zod3.z.object({
13696
- start: import_zod3.z.number().describe("The start time in milliseconds"),
13697
- end: import_zod3.z.number().describe("The end time in milliseconds")
13698
- }).describe("Timestamp containing a start and end property in milliseconds")
13699
- ).describe("The timestamp of the of the key phrase")
13700
- })
13701
- ).describe("A temporally-sequential array of Key Phrases")
13702
- }).describe(
13703
- "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"
13704
- ).or(import_zod3.z.null()).optional().describe(
13705
- "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"
14423
+ language_detection_options: import_zod3.z.object({
14424
+ expected_languages: import_zod3.z.array(import_zod3.z.string()).optional().describe(
14425
+ '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.'
14426
+ ),
14427
+ fallback_language: import_zod3.z.string().default(deleteTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault).describe(
14428
+ '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'
14429
+ ),
14430
+ code_switching: import_zod3.z.boolean().optional().describe(
14431
+ "Whether [code switching](/docs/speech-to-text/pre-recorded-audio/code-switching) should be detected.\n"
14432
+ ),
14433
+ code_switching_confidence_threshold: import_zod3.z.number().default(
14434
+ deleteTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault
14435
+ ).describe(
14436
+ "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"
14437
+ )
14438
+ }).optional().describe(
14439
+ "Specify options for [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection)."
13706
14440
  ),
13707
- audio_start_from: import_zod3.z.number().nullish().describe(
13708
- "The point in time, in milliseconds, in the file at which the transcription was started"
14441
+ multichannel: import_zod3.z.boolean().nullish().describe(
14442
+ "Whether [Multichannel transcription](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) was enabled in the transcription request, either true or false"
13709
14443
  ),
13710
- audio_end_at: import_zod3.z.number().nullish().describe(
13711
- "The point in time, in milliseconds, in the file at which the transcription was terminated"
14444
+ prompt: import_zod3.z.string().optional().describe(
14445
+ "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"
13712
14446
  ),
13713
- word_boost: import_zod3.z.array(import_zod3.z.string()).optional().describe("The list of custom vocabulary to boost transcription probability for"),
13714
- boost_param: import_zod3.z.string().nullish().describe("The word boost parameter value"),
13715
- filter_profanity: import_zod3.z.boolean().nullish().describe(
13716
- "Whether [Profanity Filtering](https://www.assemblyai.com/docs/models/speech-recognition#profanity-filtering) is enabled, either true or false"
14447
+ punctuate: import_zod3.z.boolean().nullish().describe(
14448
+ "Whether [Automatic Punctuation](https://www.assemblyai.com/docs/pre-recorded-audio) is enabled, either true or false"
13717
14449
  ),
13718
14450
  redact_pii: import_zod3.z.boolean().describe(
13719
- "Whether [PII Redaction](https://www.assemblyai.com/docs/models/pii-redaction) is enabled, either true or false"
14451
+ "Whether [PII Redaction](https://www.assemblyai.com/docs/pii-redaction) is enabled, either true or false"
13720
14452
  ),
13721
14453
  redact_pii_audio: import_zod3.z.boolean().nullish().describe(
13722
- "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"
14454
+ "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"
14455
+ ),
14456
+ redact_pii_audio_options: import_zod3.z.object({
14457
+ return_redacted_no_speech_audio: import_zod3.z.boolean().optional().describe(
14458
+ "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`."
14459
+ ),
14460
+ override_audio_redaction_method: import_zod3.z.enum(["silence"]).optional().describe(
14461
+ "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."
14462
+ )
14463
+ }).optional().describe(
14464
+ "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"
13723
14465
  ),
13724
14466
  redact_pii_audio_quality: import_zod3.z.enum(["mp3", "wav"]).describe(
13725
- "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."
14467
+ "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."
13726
14468
  ).or(import_zod3.z.null()).optional().describe(
13727
- "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"
14469
+ "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"
13728
14470
  ),
13729
14471
  redact_pii_policies: import_zod3.z.array(
13730
14472
  import_zod3.z.enum([
@@ -13774,271 +14516,256 @@ var deleteTranscriptResponse = import_zod3.z.object({
13774
14516
  "zodiac_sign"
13775
14517
  ]).describe("The type of PII to redact")
13776
14518
  ).nullish().describe(
13777
- "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"
14519
+ "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"
13778
14520
  ),
13779
14521
  redact_pii_sub: import_zod3.z.enum(["entity_name", "hash"]).optional().describe(
13780
- '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.'
14522
+ "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."
14523
+ ),
14524
+ sentiment_analysis: import_zod3.z.boolean().nullish().describe(
14525
+ "Whether [Sentiment Analysis](https://www.assemblyai.com/docs/speech-understanding/sentiment-analysis) is enabled, can be true or false"
14526
+ ),
14527
+ sentiment_analysis_results: import_zod3.z.array(
14528
+ import_zod3.z.object({
14529
+ text: import_zod3.z.string().describe("The transcript of the sentence"),
14530
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, of the sentence"),
14531
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, of the sentence"),
14532
+ sentiment: import_zod3.z.enum(["POSITIVE", "NEUTRAL", "NEGATIVE"]),
14533
+ confidence: import_zod3.z.number().describe(
14534
+ "The confidence score for the detected sentiment of the sentence, from 0 to 1"
14535
+ ),
14536
+ channel: import_zod3.z.string().nullish().describe(
14537
+ "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
14538
+ ),
14539
+ speaker: import_zod3.z.string().nullable().describe(
14540
+ "The speaker of the sentence if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
14541
+ )
14542
+ }).describe("The result of the Sentiment Analysis model")
14543
+ ).nullish().describe(
14544
+ "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"
13781
14545
  ),
13782
14546
  speaker_labels: import_zod3.z.boolean().nullish().describe(
13783
- "Whether [Speaker diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, can be true or false"
14547
+ "Whether [Speaker diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, can be true or false"
13784
14548
  ),
13785
14549
  speakers_expected: import_zod3.z.number().nullish().describe(
13786
- "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."
14550
+ "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."
13787
14551
  ),
13788
- content_safety: import_zod3.z.boolean().nullish().describe(
13789
- "Whether [Content Moderation](https://www.assemblyai.com/docs/models/content-moderation) is enabled, can be true or false"
14552
+ speech_model_used: import_zod3.z.string().optional().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."
13790
14554
  ),
13791
- content_safety_labels: import_zod3.z.object({
13792
- status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
13793
- results: import_zod3.z.array(
13794
- import_zod3.z.object({
13795
- text: import_zod3.z.string().describe("The transcript of the section flagged by the Content Moderation model"),
13796
- labels: import_zod3.z.array(
13797
- import_zod3.z.object({
13798
- label: import_zod3.z.string().describe("The label of the sensitive topic"),
13799
- confidence: import_zod3.z.number().min(
13800
- deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMin
13801
- ).max(
13802
- deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemConfidenceMax
13803
- ).describe("The confidence score for the topic being discussed, from 0 to 1"),
13804
- severity: import_zod3.z.number().min(
13805
- deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMin
13806
- ).max(
13807
- deleteTranscriptResponseContentSafetyLabelsResultsItemLabelsItemSeverityMax
13808
- ).describe("How severely the topic is discussed in the section, from 0 to 1")
13809
- })
13810
- ).describe(
13811
- "An array of safety labels, one per sensitive topic that was detected in the section"
14555
+ speech_models: import_zod3.z.array(
14556
+ import_zod3.z.string().describe(
14557
+ "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."
14558
+ )
14559
+ ).nullish().describe(
14560
+ "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"
14561
+ ),
14562
+ speech_threshold: import_zod3.z.number().nullish().describe(
14563
+ "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"
14564
+ ),
14565
+ speech_understanding: import_zod3.z.object({
14566
+ request: import_zod3.z.object({
14567
+ translation: import_zod3.z.object({
14568
+ target_languages: import_zod3.z.array(import_zod3.z.string()).describe(
14569
+ 'List of target language codes (e.g., `["es", "de"]`). See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for supported languages.'
13812
14570
  ),
13813
- sentences_idx_start: import_zod3.z.number().describe("The sentence index at which the section begins"),
13814
- sentences_idx_end: import_zod3.z.number().describe("The sentence index at which the section ends"),
13815
- timestamp: import_zod3.z.object({
13816
- start: import_zod3.z.number().describe("The start time in milliseconds"),
13817
- end: import_zod3.z.number().describe("The end time in milliseconds")
13818
- }).describe("Timestamp containing a start and end property in milliseconds")
14571
+ formal: import_zod3.z.boolean().default(deleteTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault).describe(
14572
+ "Use formal language style. See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for more details."
14573
+ ),
14574
+ match_original_utterance: import_zod3.z.boolean().optional().describe(
14575
+ "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."
14576
+ )
13819
14577
  })
13820
- ).describe("An array of results for the Content Moderation model"),
13821
- summary: import_zod3.z.record(
13822
- import_zod3.z.string(),
13823
- import_zod3.z.number().min(deleteTranscriptResponseContentSafetyLabelsSummaryMinOne).max(deleteTranscriptResponseContentSafetyLabelsSummaryMaxOne).describe(
13824
- 'A confidence score for the presence of the sensitive topic "topic" across the entire audio file'
14578
+ }).describe(
14579
+ "Request body for [Translation](https://www.assemblyai.com/docs/speech-understanding/translation)."
14580
+ ).or(
14581
+ import_zod3.z.object({
14582
+ speaker_identification: import_zod3.z.object({
14583
+ speaker_type: import_zod3.z.enum(["role", "name"]).describe(
14584
+ "Type of speaker identification. See [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification) for details on each type."
14585
+ ),
14586
+ known_values: import_zod3.z.array(import_zod3.z.string()).optional().describe(
14587
+ 'Required if speaker_type is "role". Each value must be 35 characters or less.'
14588
+ ),
14589
+ speakers: import_zod3.z.array(
14590
+ import_zod3.z.object({
14591
+ role: import_zod3.z.string().optional().describe(
14592
+ 'The role of the speaker. Required when `speaker_type` is "role".'
14593
+ ),
14594
+ name: import_zod3.z.string().optional().describe(
14595
+ 'The name of the speaker. Required when `speaker_type` is "name".'
14596
+ ),
14597
+ description: import_zod3.z.string().optional().describe(
14598
+ "A description of the speaker to help the model identify them based on conversational context."
14599
+ )
14600
+ })
14601
+ ).optional().describe(
14602
+ "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."
14603
+ )
14604
+ })
14605
+ }).describe(
14606
+ "Request body for [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification)."
13825
14607
  )
13826
- ).describe(
13827
- "A summary of the Content Moderation confidence results for the entire audio file"
13828
- ),
13829
- severity_score_summary: import_zod3.z.record(
13830
- import_zod3.z.string(),
14608
+ ).or(
14609
+ import_zod3.z.object({
14610
+ custom_formatting: import_zod3.z.object({
14611
+ date: import_zod3.z.string().optional().describe(
14612
+ 'Date format pattern (e.g., `"mm/dd/yyyy"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
14613
+ ),
14614
+ phone_number: import_zod3.z.string().optional().describe(
14615
+ 'Phone number format pattern (e.g., `"(xxx)xxx-xxxx"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
14616
+ ),
14617
+ email: import_zod3.z.string().optional().describe(
14618
+ 'Email format pattern (e.g., `"username@domain.com"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
14619
+ )
14620
+ })
14621
+ }).describe(
14622
+ "Request body for [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting)."
14623
+ )
14624
+ ).optional(),
14625
+ response: import_zod3.z.object({
14626
+ translation: import_zod3.z.object({
14627
+ status: import_zod3.z.string().optional()
14628
+ }).optional()
14629
+ }).or(
13831
14630
  import_zod3.z.object({
13832
- low: import_zod3.z.number().min(deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMin).max(deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryLowMax),
13833
- medium: import_zod3.z.number().min(deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMin).max(deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryMediumMax),
13834
- high: import_zod3.z.number().min(deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMin).max(deleteTranscriptResponseContentSafetyLabelsSeverityScoreSummaryHighMax)
14631
+ speaker_identification: import_zod3.z.object({
14632
+ mapping: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional().describe(
14633
+ 'A mapping of the original generic speaker labels (e.g., "A", "B") to the identified speaker names or roles.'
14634
+ ),
14635
+ status: import_zod3.z.string().optional()
14636
+ }).optional()
13835
14637
  })
13836
- ).describe(
13837
- "A summary of the Content Moderation severity results for the entire audio file"
13838
- )
13839
- }).describe(
13840
- "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"
13841
- ).or(import_zod3.z.null()).optional().describe(
13842
- "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"
13843
- ),
13844
- iab_categories: import_zod3.z.boolean().nullish().describe(
13845
- "Whether [Topic Detection](https://www.assemblyai.com/docs/models/topic-detection) is enabled, can be true or false"
13846
- ),
13847
- iab_categories_result: import_zod3.z.object({
13848
- status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
13849
- results: import_zod3.z.array(
14638
+ ).or(
13850
14639
  import_zod3.z.object({
13851
- text: import_zod3.z.string().describe("The text in the transcript in which a detected topic occurs"),
13852
- labels: import_zod3.z.array(
13853
- import_zod3.z.object({
13854
- relevance: import_zod3.z.number().min(
13855
- deleteTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMin
13856
- ).max(
13857
- deleteTranscriptResponseIabCategoriesResultResultsItemLabelsItemRelevanceMax
13858
- ).describe("How relevant the detected topic is of a detected topic"),
13859
- label: import_zod3.z.string().describe(
13860
- "The IAB taxonomical label for the label of the detected topic, where > denotes supertopic/subtopic relationship"
13861
- )
13862
- })
13863
- ).optional().describe("An array of detected topics in the text"),
13864
- timestamp: import_zod3.z.object({
13865
- start: import_zod3.z.number().describe("The start time in milliseconds"),
13866
- end: import_zod3.z.number().describe("The end time in milliseconds")
13867
- }).optional().describe("Timestamp containing a start and end property in milliseconds")
13868
- }).describe("The result of the topic detection model")
13869
- ).describe("An array of results for the Topic Detection model"),
13870
- summary: import_zod3.z.record(
13871
- import_zod3.z.string(),
13872
- import_zod3.z.number().min(deleteTranscriptResponseIabCategoriesResultSummaryMinOne).max(deleteTranscriptResponseIabCategoriesResultSummaryMaxOne)
13873
- ).describe("The overall relevance of topic to the entire audio file")
13874
- }).describe(
13875
- "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"
13876
- ).or(import_zod3.z.null()).optional().describe(
13877
- "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"
13878
- ),
13879
- custom_spelling: import_zod3.z.array(
13880
- import_zod3.z.object({
13881
- from: import_zod3.z.array(import_zod3.z.string().describe("Word or phrase to replace")).describe("Words or phrases to replace"),
13882
- to: import_zod3.z.string().describe("Word to replace with")
13883
- }).describe(
13884
- "Object containing words or phrases to replace, and the word or phrase to replace with"
13885
- )
13886
- ).nullish().describe("Customize how words are spelled and formatted using to and from values"),
13887
- keyterms_prompt: import_zod3.z.array(import_zod3.z.string()).optional().describe(
13888
- "Improve accuracy with up to 1000 domain-specific words or phrases (maximum 6 words per phrase).\n"
14640
+ custom_formatting: import_zod3.z.object({
14641
+ mapping: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional(),
14642
+ formatted_text: import_zod3.z.string().optional()
14643
+ }).optional()
14644
+ })
14645
+ ).optional()
14646
+ }).optional().describe(
14647
+ "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"
13889
14648
  ),
13890
- prompt: import_zod3.z.string().optional().describe("This parameter does not currently have any functionality attached to it."),
13891
- auto_chapters: import_zod3.z.boolean().nullish().describe(
13892
- "Whether [Auto Chapters](https://www.assemblyai.com/docs/models/auto-chapters) is enabled, can be true or false"
14649
+ status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).describe(
14650
+ "The status of your transcript. Possible values are queued, processing, completed, or error."
13893
14651
  ),
13894
- chapters: import_zod3.z.array(
13895
- import_zod3.z.object({
13896
- gist: import_zod3.z.string().describe(
13897
- "An ultra-short summary (just a few words) of the content spoken in the chapter"
13898
- ),
13899
- headline: import_zod3.z.string().describe("A single sentence summary of the content spoken during the chapter"),
13900
- summary: import_zod3.z.string().describe("A one paragraph summary of the content spoken during the chapter"),
13901
- start: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter"),
13902
- end: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter")
13903
- }).describe("Chapter of the audio file")
13904
- ).nullish().describe("An array of temporally sequential chapters for the audio file"),
13905
14652
  summarization: import_zod3.z.boolean().describe(
13906
- "Whether [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled, either true or false"
14653
+ "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"
13907
14654
  ),
13908
- summary_type: import_zod3.z.string().nullish().describe(
13909
- "The type of summary generated, if [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled"
14655
+ summary: import_zod3.z.string().nullish().describe(
14656
+ "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."
13910
14657
  ),
13911
14658
  summary_model: import_zod3.z.string().nullish().describe(
13912
- "The Summarization model used to generate the summary,\nif [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled\n"
14659
+ "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"
13913
14660
  ),
13914
- summary: import_zod3.z.string().nullish().describe(
13915
- "The generated summary of the media file, if [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled"
14661
+ summary_type: import_zod3.z.string().nullish().describe(
14662
+ "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."
13916
14663
  ),
13917
- custom_topics: import_zod3.z.boolean().nullish().describe("Whether custom topics is enabled, either true or false"),
13918
- topics: import_zod3.z.array(import_zod3.z.string()).optional().describe("The list of custom topics provided if custom topics is enabled"),
13919
- sentiment_analysis: import_zod3.z.boolean().nullish().describe(
13920
- "Whether [Sentiment Analysis](https://www.assemblyai.com/docs/models/sentiment-analysis) is enabled, can be true or false"
14664
+ remove_audio_tags: import_zod3.z.enum(["all"]).describe(
14665
+ "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"
14666
+ ).or(import_zod3.z.null()).optional().describe(
14667
+ "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"
13921
14668
  ),
13922
- sentiment_analysis_results: import_zod3.z.array(
14669
+ temperature: import_zod3.z.number().nullish().describe(
14670
+ "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"
14671
+ ),
14672
+ text: import_zod3.z.string().nullish().describe("The textual transcript of your media file"),
14673
+ throttled: import_zod3.z.boolean().nullish().describe(
14674
+ "True while a request is throttled and false when a request is no longer throttled"
14675
+ ),
14676
+ utterances: import_zod3.z.array(
13923
14677
  import_zod3.z.object({
13924
- text: import_zod3.z.string().describe("The transcript of the sentence"),
13925
- start: import_zod3.z.number().describe("The starting time, in milliseconds, of the sentence"),
13926
- end: import_zod3.z.number().describe("The ending time, in milliseconds, of the sentence"),
13927
- sentiment: import_zod3.z.enum(["POSITIVE", "NEUTRAL", "NEGATIVE"]),
13928
- confidence: import_zod3.z.number().min(deleteTranscriptResponseSentimentAnalysisResultsItemConfidenceMin).max(deleteTranscriptResponseSentimentAnalysisResultsItemConfidenceMax).describe(
13929
- "The confidence score for the detected sentiment of the sentence, from 0 to 1"
13930
- ),
14678
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this utterance"),
14679
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, of the utterance in the audio file"),
14680
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, of the utterance in the audio file"),
14681
+ text: import_zod3.z.string().describe("The text for this utterance"),
14682
+ words: import_zod3.z.array(
14683
+ import_zod3.z.object({
14684
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
14685
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
14686
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
14687
+ text: import_zod3.z.string().describe("The text of the word"),
14688
+ channel: import_zod3.z.string().nullish().describe(
14689
+ "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
14690
+ ),
14691
+ speaker: import_zod3.z.string().nullable().describe(
14692
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
14693
+ )
14694
+ })
14695
+ ).describe("The words in the utterance."),
13931
14696
  channel: import_zod3.z.string().nullish().describe(
13932
14697
  "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13933
14698
  ),
13934
- speaker: import_zod3.z.string().nullable().describe(
13935
- "The speaker of the sentence if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
14699
+ speaker: import_zod3.z.string().describe(
14700
+ '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.'
14701
+ ),
14702
+ translated_texts: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional().describe(
14703
+ 'Translations keyed by language code (e.g., `{"es": "Texto traducido", "de": "\xDCbersetzter Text"}`). Only present when `match_original_utterance` is enabled with translation.'
13936
14704
  )
13937
- }).describe("The result of the Sentiment Analysis model")
14705
+ })
13938
14706
  ).nullish().describe(
13939
- "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"
14707
+ "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"
13940
14708
  ),
13941
- entity_detection: import_zod3.z.boolean().nullish().describe(
13942
- "Whether [Entity Detection](https://www.assemblyai.com/docs/models/entity-detection) is enabled, can be true or false"
14709
+ webhook_auth: import_zod3.z.boolean().describe(
14710
+ "Whether [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) authentication details were provided"
13943
14711
  ),
13944
- entities: import_zod3.z.array(
14712
+ webhook_auth_header_name: import_zod3.z.string().nullish().describe(
14713
+ "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"
14714
+ ),
14715
+ webhook_status_code: import_zod3.z.number().nullish().describe(
14716
+ "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"
14717
+ ),
14718
+ webhook_url: import_zod3.z.string().nullish().describe(
14719
+ "The URL to which we send [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) requests.\n"
14720
+ ),
14721
+ words: import_zod3.z.array(
13945
14722
  import_zod3.z.object({
13946
- entity_type: import_zod3.z.enum([
13947
- "account_number",
13948
- "banking_information",
13949
- "blood_type",
13950
- "credit_card_cvv",
13951
- "credit_card_expiration",
13952
- "credit_card_number",
13953
- "date",
13954
- "date_interval",
13955
- "date_of_birth",
13956
- "drivers_license",
13957
- "drug",
13958
- "duration",
13959
- "email_address",
13960
- "event",
13961
- "filename",
13962
- "gender_sexuality",
13963
- "healthcare_number",
13964
- "injury",
13965
- "ip_address",
13966
- "language",
13967
- "location",
13968
- "marital_status",
13969
- "medical_condition",
13970
- "medical_process",
13971
- "money_amount",
13972
- "nationality",
13973
- "number_sequence",
13974
- "occupation",
13975
- "organization",
13976
- "passport_number",
13977
- "password",
13978
- "person_age",
13979
- "person_name",
13980
- "phone_number",
13981
- "physical_attribute",
13982
- "political_affiliation",
13983
- "religion",
13984
- "statistics",
13985
- "time",
13986
- "url",
13987
- "us_social_security_number",
13988
- "username",
13989
- "vehicle_id",
13990
- "zodiac_sign"
13991
- ]).describe("The type of entity for the detected entity"),
13992
- text: import_zod3.z.string().describe("The text for the detected entity"),
13993
- start: import_zod3.z.number().describe(
13994
- "The starting time, in milliseconds, at which the detected entity appears in the audio file"
14723
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
14724
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
14725
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
14726
+ text: import_zod3.z.string().describe("The text of the word"),
14727
+ channel: import_zod3.z.string().nullish().describe(
14728
+ "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13995
14729
  ),
13996
- end: import_zod3.z.number().describe(
13997
- "The ending time, in milliseconds, for the detected entity in the audio file"
14730
+ speaker: import_zod3.z.string().nullable().describe(
14731
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
13998
14732
  )
13999
- }).describe("A detected entity")
14733
+ })
14000
14734
  ).nullish().describe(
14001
- "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"
14002
- ),
14003
- speech_threshold: import_zod3.z.number().min(deleteTranscriptResponseSpeechThresholdMin).max(deleteTranscriptResponseSpeechThresholdMax).nullish().describe(
14004
- "Defaults to null. Reject audio files that contain less than this fraction of speech.\nValid values are in the range [0, 1] inclusive.\n"
14005
- ),
14006
- throttled: import_zod3.z.boolean().nullish().describe(
14007
- "True while a request is throttled and false when a request is no longer throttled"
14008
- ),
14009
- error: import_zod3.z.string().optional().describe("Error message of why the transcript failed"),
14010
- language_model: import_zod3.z.string().describe("The language model that was used for the transcript"),
14011
- acoustic_model: import_zod3.z.string().describe("The acoustic model that was used for the transcript")
14735
+ "An array of temporally-sequential word objects, one for each word in the transcript.\n"
14736
+ ),
14737
+ acoustic_model: import_zod3.z.string().describe("This parameter does not currently have any functionality attached to it."),
14738
+ custom_topics: import_zod3.z.boolean().nullish().describe("This parameter does not currently have any functionality attached to it."),
14739
+ language_model: import_zod3.z.string().describe("This parameter does not currently have any functionality attached to it."),
14740
+ speech_model: import_zod3.z.string().describe(
14741
+ "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."
14742
+ ).or(import_zod3.z.null()).describe(
14743
+ "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"
14744
+ ),
14745
+ speed_boost: import_zod3.z.boolean().nullish().describe("This parameter does not currently have any functionality attached to it."),
14746
+ topics: import_zod3.z.array(import_zod3.z.string()).optional().describe("This parameter does not currently have any functionality attached to it."),
14747
+ translated_texts: import_zod3.z.object({
14748
+ language_code: import_zod3.z.string().optional().describe("Translated text for this language code")
14749
+ }).optional().describe(
14750
+ "Translated text keyed by language code. See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for more details."
14751
+ )
14012
14752
  }).describe("A transcript object");
14013
- var getSubtitlesParams = import_zod3.z.object({
14014
- transcript_id: import_zod3.z.string().describe("ID of the transcript"),
14015
- subtitle_format: import_zod3.z.enum(["srt", "vtt"]).describe("The format of the captions")
14016
- });
14017
- var getSubtitlesQueryParams = import_zod3.z.object({
14018
- chars_per_caption: import_zod3.z.number().optional().describe("The maximum number of characters per caption")
14019
- });
14020
14753
  var getTranscriptSentencesParams = import_zod3.z.object({
14021
14754
  transcript_id: import_zod3.z.string().describe("ID of the transcript")
14022
14755
  });
14023
- var getTranscriptSentencesResponseConfidenceMin = 0;
14024
- var getTranscriptSentencesResponseConfidenceMax = 1;
14025
- var getTranscriptSentencesResponseSentencesItemConfidenceMin = 0;
14026
- var getTranscriptSentencesResponseSentencesItemConfidenceMax = 1;
14027
- var getTranscriptSentencesResponseSentencesItemWordsItemConfidenceMin = 0;
14028
- var getTranscriptSentencesResponseSentencesItemWordsItemConfidenceMax = 1;
14029
14756
  var getTranscriptSentencesResponse = import_zod3.z.object({
14030
14757
  id: import_zod3.z.string().uuid().describe("The unique identifier for the transcript"),
14031
- confidence: import_zod3.z.number().min(getTranscriptSentencesResponseConfidenceMin).max(getTranscriptSentencesResponseConfidenceMax).describe("The confidence score for the transcript"),
14758
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript"),
14032
14759
  audio_duration: import_zod3.z.number().describe("The duration of the audio file in seconds"),
14033
14760
  sentences: import_zod3.z.array(
14034
14761
  import_zod3.z.object({
14035
14762
  text: import_zod3.z.string().describe("The transcript of the sentence"),
14036
14763
  start: import_zod3.z.number().describe("The starting time, in milliseconds, for the sentence"),
14037
14764
  end: import_zod3.z.number().describe("The ending time, in milliseconds, for the sentence"),
14038
- confidence: import_zod3.z.number().min(getTranscriptSentencesResponseSentencesItemConfidenceMin).max(getTranscriptSentencesResponseSentencesItemConfidenceMax).describe("The confidence score for the transcript of this sentence"),
14765
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this sentence"),
14039
14766
  words: import_zod3.z.array(
14040
14767
  import_zod3.z.object({
14041
- confidence: import_zod3.z.number().min(getTranscriptSentencesResponseSentencesItemWordsItemConfidenceMin).max(getTranscriptSentencesResponseSentencesItemWordsItemConfidenceMax).describe("The confidence score for the transcript of this word"),
14768
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
14042
14769
  start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
14043
14770
  end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
14044
14771
  text: import_zod3.z.string().describe("The text of the word"),
@@ -14046,7 +14773,7 @@ var getTranscriptSentencesResponse = import_zod3.z.object({
14046
14773
  "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
14047
14774
  ),
14048
14775
  speaker: import_zod3.z.string().nullable().describe(
14049
- "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
14776
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
14050
14777
  )
14051
14778
  })
14052
14779
  ).describe("An array of words in the sentence"),
@@ -14054,7 +14781,7 @@ var getTranscriptSentencesResponse = import_zod3.z.object({
14054
14781
  "The channel of the sentence. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
14055
14782
  ),
14056
14783
  speaker: import_zod3.z.string().nullable().describe(
14057
- "The speaker of the sentence if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
14784
+ "The speaker of the sentence if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
14058
14785
  )
14059
14786
  })
14060
14787
  ).describe("An array of sentences in the transcript")
@@ -14062,25 +14789,19 @@ var getTranscriptSentencesResponse = import_zod3.z.object({
14062
14789
  var getTranscriptParagraphsParams = import_zod3.z.object({
14063
14790
  transcript_id: import_zod3.z.string().describe("ID of the transcript")
14064
14791
  });
14065
- var getTranscriptParagraphsResponseConfidenceMin = 0;
14066
- var getTranscriptParagraphsResponseConfidenceMax = 1;
14067
- var getTranscriptParagraphsResponseParagraphsItemConfidenceMin = 0;
14068
- var getTranscriptParagraphsResponseParagraphsItemConfidenceMax = 1;
14069
- var getTranscriptParagraphsResponseParagraphsItemWordsItemConfidenceMin = 0;
14070
- var getTranscriptParagraphsResponseParagraphsItemWordsItemConfidenceMax = 1;
14071
14792
  var getTranscriptParagraphsResponse = import_zod3.z.object({
14072
14793
  id: import_zod3.z.string().uuid().describe("The unique identifier of your transcript"),
14073
- confidence: import_zod3.z.number().min(getTranscriptParagraphsResponseConfidenceMin).max(getTranscriptParagraphsResponseConfidenceMax).describe("The confidence score for the transcript"),
14794
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript"),
14074
14795
  audio_duration: import_zod3.z.number().describe("The duration of the audio file in seconds"),
14075
14796
  paragraphs: import_zod3.z.array(
14076
14797
  import_zod3.z.object({
14077
14798
  text: import_zod3.z.string().describe("The transcript of the paragraph"),
14078
14799
  start: import_zod3.z.number().describe("The starting time, in milliseconds, of the paragraph"),
14079
14800
  end: import_zod3.z.number().describe("The ending time, in milliseconds, of the paragraph"),
14080
- confidence: import_zod3.z.number().min(getTranscriptParagraphsResponseParagraphsItemConfidenceMin).max(getTranscriptParagraphsResponseParagraphsItemConfidenceMax).describe("The confidence score for the transcript of this paragraph"),
14801
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this paragraph"),
14081
14802
  words: import_zod3.z.array(
14082
14803
  import_zod3.z.object({
14083
- confidence: import_zod3.z.number().min(getTranscriptParagraphsResponseParagraphsItemWordsItemConfidenceMin).max(getTranscriptParagraphsResponseParagraphsItemWordsItemConfidenceMax).describe("The confidence score for the transcript of this word"),
14804
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
14084
14805
  start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
14085
14806
  end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
14086
14807
  text: import_zod3.z.string().describe("The text of the word"),
@@ -14088,13 +14809,28 @@ var getTranscriptParagraphsResponse = import_zod3.z.object({
14088
14809
  "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
14089
14810
  ),
14090
14811
  speaker: import_zod3.z.string().nullable().describe(
14091
- "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/models/speaker-diarization) is enabled, else null"
14812
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
14092
14813
  )
14093
14814
  })
14094
14815
  ).describe("An array of words in the paragraph")
14095
14816
  })
14096
14817
  ).describe("An array of paragraphs in the transcript")
14097
14818
  });
14819
+ var getSubtitlesParams = import_zod3.z.object({
14820
+ transcript_id: import_zod3.z.string().describe("ID of the transcript"),
14821
+ subtitle_format: import_zod3.z.enum(["srt", "vtt"]).describe("The format of the captions")
14822
+ });
14823
+ var getSubtitlesQueryParams = import_zod3.z.object({
14824
+ chars_per_caption: import_zod3.z.number().optional().describe("The maximum number of characters per caption")
14825
+ });
14826
+ var getSubtitlesResponse = import_zod3.z.object({});
14827
+ var getRedactedAudioParams = import_zod3.z.object({
14828
+ transcript_id: import_zod3.z.string().describe("ID of the transcript")
14829
+ });
14830
+ var getRedactedAudioResponse = import_zod3.z.object({
14831
+ status: import_zod3.z.enum(["redacted_audio_ready"]).describe("The status of the redacted audio"),
14832
+ redacted_audio_url: import_zod3.z.string().describe("The URL of the redacted audio file")
14833
+ });
14098
14834
  var wordSearchParams = import_zod3.z.object({
14099
14835
  transcript_id: import_zod3.z.string().describe("ID of the transcript")
14100
14836
  });
@@ -14111,7 +14847,7 @@ var wordSearchResponse = import_zod3.z.object({
14111
14847
  text: import_zod3.z.string().describe("The matched word"),
14112
14848
  count: import_zod3.z.number().describe("The total amount of times the word is in the transcript"),
14113
14849
  timestamps: import_zod3.z.array(
14114
- import_zod3.z.array(import_zod3.z.number().describe("Timestamp in milliseconds")).describe(
14850
+ import_zod3.z.array(import_zod3.z.number()).describe(
14115
14851
  "An array of timestamps structured as [`start_time`, `end_time`] in milliseconds"
14116
14852
  )
14117
14853
  ).describe("An array of timestamps"),
@@ -14121,205 +14857,6 @@ var wordSearchResponse = import_zod3.z.object({
14121
14857
  })
14122
14858
  ).describe("The matches of the search")
14123
14859
  });
14124
- var getRedactedAudioParams = import_zod3.z.object({
14125
- transcript_id: import_zod3.z.string().describe("ID of the transcript")
14126
- });
14127
- var getRedactedAudioResponse = import_zod3.z.object({
14128
- status: import_zod3.z.enum(["redacted_audio_ready"]).describe("The status of the redacted audio"),
14129
- redacted_audio_url: import_zod3.z.string().describe("The URL of the redacted audio file")
14130
- });
14131
- var createTemporaryTokenBodyExpiresInMin = 60;
14132
- var createTemporaryTokenBody = import_zod3.z.object({
14133
- expires_in: import_zod3.z.number().min(createTemporaryTokenBodyExpiresInMin).describe("The amount of time until the token expires in seconds")
14134
- });
14135
- var createTemporaryTokenResponse = import_zod3.z.object({
14136
- token: import_zod3.z.string().describe("The temporary authentication token for Streaming Speech-to-Text")
14137
- });
14138
- var lemurTaskBodyFinalModelDefault = "default";
14139
- var lemurTaskBodyMaxOutputSizeDefault = 2e3;
14140
- var lemurTaskBodyTemperatureDefault = 0;
14141
- var lemurTaskBodyTemperatureMin = 0;
14142
- var lemurTaskBodyTemperatureMax = 1;
14143
- var lemurTaskBody = import_zod3.z.object({
14144
- prompt: import_zod3.z.string().describe(
14145
- "Your text to prompt the model to produce a desired output, including any context you want to pass into the model."
14146
- )
14147
- }).and(
14148
- import_zod3.z.object({
14149
- transcript_ids: import_zod3.z.array(import_zod3.z.string().uuid()).optional().describe(
14150
- "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"
14151
- ),
14152
- input_text: import_zod3.z.string().optional().describe(
14153
- "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"
14154
- ),
14155
- 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."),
14156
- final_model: import_zod3.z.enum([
14157
- "anthropic/claude-3-5-sonnet",
14158
- "anthropic/claude-3-opus",
14159
- "anthropic/claude-3-haiku"
14160
- ]).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"),
14161
- max_output_size: import_zod3.z.number().default(lemurTaskBodyMaxOutputSizeDefault).describe("Max output size in tokens."),
14162
- temperature: import_zod3.z.number().min(lemurTaskBodyTemperatureMin).max(lemurTaskBodyTemperatureMax).optional().describe(
14163
- "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"
14164
- )
14165
- })
14166
- );
14167
- var lemurTaskResponseUsageInputTokensMin = 0;
14168
- var lemurTaskResponseUsageOutputTokensMin = 0;
14169
- var lemurTaskResponse = import_zod3.z.object({
14170
- response: import_zod3.z.string().describe("The response generated by LeMUR.")
14171
- }).and(
14172
- import_zod3.z.object({
14173
- request_id: import_zod3.z.string().uuid().describe("The ID of the LeMUR request"),
14174
- usage: import_zod3.z.object({
14175
- input_tokens: import_zod3.z.number().min(lemurTaskResponseUsageInputTokensMin).describe("The number of input tokens used by the model"),
14176
- output_tokens: import_zod3.z.number().min(lemurTaskResponseUsageOutputTokensMin).describe("The number of output tokens generated by the model")
14177
- }).describe("The usage numbers for the LeMUR request")
14178
- })
14179
- );
14180
- var lemurSummaryBodyFinalModelDefault = "default";
14181
- var lemurSummaryBodyMaxOutputSizeDefault = 2e3;
14182
- var lemurSummaryBodyTemperatureDefault = 0;
14183
- var lemurSummaryBodyTemperatureMin = 0;
14184
- var lemurSummaryBodyTemperatureMax = 1;
14185
- var lemurSummaryBody = import_zod3.z.object({
14186
- transcript_ids: import_zod3.z.array(import_zod3.z.string().uuid()).optional().describe(
14187
- "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"
14188
- ),
14189
- input_text: import_zod3.z.string().optional().describe(
14190
- "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"
14191
- ),
14192
- 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."),
14193
- 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"),
14194
- max_output_size: import_zod3.z.number().default(lemurSummaryBodyMaxOutputSizeDefault).describe("Max output size in tokens."),
14195
- temperature: import_zod3.z.number().min(lemurSummaryBodyTemperatureMin).max(lemurSummaryBodyTemperatureMax).optional().describe(
14196
- "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"
14197
- )
14198
- }).and(
14199
- import_zod3.z.object({
14200
- answer_format: import_zod3.z.string().optional().describe(
14201
- 'How you want the summary to be returned. This can be any text. Examples: "TLDR", "bullet points"\n'
14202
- )
14203
- })
14204
- );
14205
- var lemurSummaryResponseUsageInputTokensMin = 0;
14206
- var lemurSummaryResponseUsageOutputTokensMin = 0;
14207
- var lemurSummaryResponse = import_zod3.z.object({
14208
- response: import_zod3.z.string().describe("The response generated by LeMUR.")
14209
- }).and(
14210
- import_zod3.z.object({
14211
- request_id: import_zod3.z.string().uuid().describe("The ID of the LeMUR request"),
14212
- usage: import_zod3.z.object({
14213
- input_tokens: import_zod3.z.number().min(lemurSummaryResponseUsageInputTokensMin).describe("The number of input tokens used by the model"),
14214
- output_tokens: import_zod3.z.number().min(lemurSummaryResponseUsageOutputTokensMin).describe("The number of output tokens generated by the model")
14215
- }).describe("The usage numbers for the LeMUR request")
14216
- })
14217
- );
14218
- var lemurQuestionAnswerBodyFinalModelDefault = "default";
14219
- var lemurQuestionAnswerBodyMaxOutputSizeDefault = 2e3;
14220
- var lemurQuestionAnswerBodyTemperatureDefault = 0;
14221
- var lemurQuestionAnswerBodyTemperatureMin = 0;
14222
- var lemurQuestionAnswerBodyTemperatureMax = 1;
14223
- var lemurQuestionAnswerBody = import_zod3.z.object({
14224
- transcript_ids: import_zod3.z.array(import_zod3.z.string().uuid()).optional().describe(
14225
- "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"
14226
- ),
14227
- input_text: import_zod3.z.string().optional().describe(
14228
- "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"
14229
- ),
14230
- 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."),
14231
- 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"),
14232
- max_output_size: import_zod3.z.number().default(lemurQuestionAnswerBodyMaxOutputSizeDefault).describe("Max output size in tokens."),
14233
- temperature: import_zod3.z.number().min(lemurQuestionAnswerBodyTemperatureMin).max(lemurQuestionAnswerBodyTemperatureMax).optional().describe(
14234
- "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"
14235
- )
14236
- }).and(
14237
- import_zod3.z.object({
14238
- questions: import_zod3.z.array(
14239
- import_zod3.z.object({
14240
- question: import_zod3.z.string().describe(
14241
- "The question you wish to ask. For more complex questions use default model."
14242
- ),
14243
- context: import_zod3.z.string().or(import_zod3.z.record(import_zod3.z.string(), import_zod3.z.any())).optional().describe(
14244
- "Any context about the transcripts you wish to provide. This can be a string or any object."
14245
- ),
14246
- answer_format: import_zod3.z.string().optional().describe(
14247
- `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"
14248
- `
14249
- ),
14250
- answer_options: import_zod3.z.array(import_zod3.z.string()).optional().describe(
14251
- `What discrete options to return. Useful for precise responses. Can't be used with answer_format. Example: ["Yes", "No"]
14252
- `
14253
- )
14254
- })
14255
- ).describe("A list of questions to ask")
14256
- })
14257
- );
14258
- var lemurQuestionAnswerResponseUsageInputTokensMin = 0;
14259
- var lemurQuestionAnswerResponseUsageOutputTokensMin = 0;
14260
- var lemurQuestionAnswerResponse = import_zod3.z.object({
14261
- request_id: import_zod3.z.string().uuid().describe("The ID of the LeMUR request"),
14262
- usage: import_zod3.z.object({
14263
- input_tokens: import_zod3.z.number().min(lemurQuestionAnswerResponseUsageInputTokensMin).describe("The number of input tokens used by the model"),
14264
- output_tokens: import_zod3.z.number().min(lemurQuestionAnswerResponseUsageOutputTokensMin).describe("The number of output tokens generated by the model")
14265
- }).describe("The usage numbers for the LeMUR request")
14266
- }).and(
14267
- import_zod3.z.object({
14268
- response: import_zod3.z.array(
14269
- import_zod3.z.object({
14270
- question: import_zod3.z.string().describe("The question for LeMUR to answer"),
14271
- answer: import_zod3.z.string().describe("The answer generated by LeMUR")
14272
- }).describe("An answer generated by LeMUR and its question")
14273
- ).describe("The answers generated by LeMUR and their questions")
14274
- })
14275
- );
14276
- var getLemurResponseParams = import_zod3.z.object({
14277
- request_id: import_zod3.z.string().describe(
14278
- "The ID of the LeMUR request you previously made.\nThis would be found in the response of the original request.\n"
14279
- )
14280
- });
14281
- var getLemurResponseResponseUsageInputTokensMin = 0;
14282
- var getLemurResponseResponseUsageOutputTokensMin = 0;
14283
- var getLemurResponseResponseUsageInputTokensMinOne = 0;
14284
- var getLemurResponseResponseUsageOutputTokensMinOne = 0;
14285
- var getLemurResponseResponse = import_zod3.z.object({
14286
- response: import_zod3.z.string().describe("The response generated by LeMUR.")
14287
- }).and(
14288
- import_zod3.z.object({
14289
- request_id: import_zod3.z.string().uuid().describe("The ID of the LeMUR request"),
14290
- usage: import_zod3.z.object({
14291
- input_tokens: import_zod3.z.number().min(getLemurResponseResponseUsageInputTokensMin).describe("The number of input tokens used by the model"),
14292
- output_tokens: import_zod3.z.number().min(getLemurResponseResponseUsageOutputTokensMin).describe("The number of output tokens generated by the model")
14293
- }).describe("The usage numbers for the LeMUR request")
14294
- })
14295
- ).or(
14296
- import_zod3.z.object({
14297
- request_id: import_zod3.z.string().uuid().describe("The ID of the LeMUR request"),
14298
- usage: import_zod3.z.object({
14299
- input_tokens: import_zod3.z.number().min(getLemurResponseResponseUsageInputTokensMinOne).describe("The number of input tokens used by the model"),
14300
- output_tokens: import_zod3.z.number().min(getLemurResponseResponseUsageOutputTokensMinOne).describe("The number of output tokens generated by the model")
14301
- }).describe("The usage numbers for the LeMUR request")
14302
- }).and(
14303
- import_zod3.z.object({
14304
- response: import_zod3.z.array(
14305
- import_zod3.z.object({
14306
- question: import_zod3.z.string().describe("The question for LeMUR to answer"),
14307
- answer: import_zod3.z.string().describe("The answer generated by LeMUR")
14308
- }).describe("An answer generated by LeMUR and its question")
14309
- ).describe("The answers generated by LeMUR and their questions")
14310
- })
14311
- )
14312
- );
14313
- var purgeLemurRequestDataParams = import_zod3.z.object({
14314
- request_id: import_zod3.z.string().describe(
14315
- "The ID of the LeMUR request whose data you want to delete. This would be found in the response of the original request."
14316
- )
14317
- });
14318
- var purgeLemurRequestDataResponse = import_zod3.z.object({
14319
- request_id: import_zod3.z.string().uuid().describe("The ID of the deletion request of the LeMUR request"),
14320
- request_id_to_purge: import_zod3.z.string().uuid().describe("The ID of the LeMUR request to purge the data for"),
14321
- deleted: import_zod3.z.boolean().describe("Whether the request data was deleted")
14322
- });
14323
14860
 
14324
14861
  // src/generated/assemblyai/streaming-types.zod.ts
14325
14862
  var import_zod4 = require("zod");
@@ -14334,25 +14871,37 @@ var streamingTranscriberParams = import_zod4.z.object({
14334
14871
  enableExtraSessionInformation: import_zod4.z.boolean().optional().describe(
14335
14872
  "Set to true to receive the SessionInformation message before the session ends. Defaults to false."
14336
14873
  ),
14874
+ domain: import_zod4.z.string().optional().describe(
14875
+ '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.'
14876
+ ),
14337
14877
  endOfTurnConfidenceThreshold: import_zod4.z.number().optional().describe("From SDK v3"),
14338
14878
  minEndOfTurnSilenceWhenConfident: import_zod4.z.number().optional().describe("From SDK v3"),
14879
+ minTurnSilence: import_zod4.z.number().optional().describe("From SDK v3"),
14339
14880
  maxTurnSilence: import_zod4.z.number().optional().describe("From SDK v3"),
14340
14881
  vadThreshold: import_zod4.z.number().optional().describe("From SDK v3"),
14341
14882
  formatTurns: import_zod4.z.boolean().optional().describe("From SDK v3"),
14342
14883
  filterProfanity: import_zod4.z.boolean().optional().describe("From SDK v3"),
14343
14884
  keyterms: import_zod4.z.array(import_zod4.z.string()).optional().describe("From SDK v3"),
14344
14885
  keytermsPrompt: import_zod4.z.array(import_zod4.z.string()).optional().describe("From SDK v3"),
14345
- speechModel: import_zod4.z.enum(["universal-streaming-english", "universal-streaming-multilingual"]).optional().describe("From SDK v3"),
14886
+ prompt: import_zod4.z.string().optional().describe("From SDK v3"),
14887
+ speechModel: import_zod4.z.enum(["universal-streaming-english", "universal-streaming-multilingual"]).describe("From SDK v3"),
14346
14888
  languageDetection: import_zod4.z.boolean().optional().describe("From SDK v3"),
14347
- inactivityTimeout: import_zod4.z.number().optional().describe("From SDK v3")
14889
+ inactivityTimeout: import_zod4.z.number().optional().describe("From SDK v3"),
14890
+ speakerLabels: import_zod4.z.boolean().optional().describe("From SDK v3"),
14891
+ maxSpeakers: import_zod4.z.number().optional().describe("From SDK v3"),
14892
+ llmGateway: import_zod4.z.unknown().optional().describe("From SDK v3")
14348
14893
  });
14349
14894
  var streamingUpdateConfigParams = import_zod4.z.object({
14350
14895
  end_utterance_silence_threshold: import_zod4.z.number().min(0).max(2e4).optional().describe("The duration threshold in milliseconds"),
14351
14896
  end_of_turn_confidence_threshold: import_zod4.z.number().optional().describe("From SDK v3"),
14352
14897
  min_end_of_turn_silence_when_confident: import_zod4.z.number().optional().describe("From SDK v3"),
14898
+ min_turn_silence: import_zod4.z.number().optional().describe("From SDK v3"),
14353
14899
  max_turn_silence: import_zod4.z.number().optional().describe("From SDK v3"),
14354
14900
  vad_threshold: import_zod4.z.number().optional().describe("From SDK v3"),
14355
- format_turns: import_zod4.z.boolean().optional().describe("From SDK v3")
14901
+ format_turns: import_zod4.z.boolean().optional().describe("From SDK v3"),
14902
+ keyterms_prompt: import_zod4.z.array(import_zod4.z.string()).optional().describe("From SDK v3"),
14903
+ prompt: import_zod4.z.string().optional().describe("From SDK v3"),
14904
+ filter_profanity: import_zod4.z.boolean().optional().describe("From SDK v3")
14356
14905
  });
14357
14906
 
14358
14907
  // src/generated/gladia/api/gladiaControlAPI.zod.ts
@@ -40376,7 +40925,6 @@ var deleteTranscriptByIdResponse = import_zod13.z.any();
40376
40925
  createOpenAIWhisperAdapter,
40377
40926
  createSonioxAdapter,
40378
40927
  createSpeechmaticsAdapter,
40379
- createTemporaryToken,
40380
40928
  createTranscript,
40381
40929
  createTranscription,
40382
40930
  createVoiceRouter,