voice-router-dev 0.8.3 → 0.8.4

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/dist/index.js CHANGED
@@ -182,7 +182,6 @@ __export(src_exports, {
182
182
  createOpenAIWhisperAdapter: () => createOpenAIWhisperAdapter,
183
183
  createSonioxAdapter: () => createSonioxAdapter,
184
184
  createSpeechmaticsAdapter: () => createSpeechmaticsAdapter,
185
- createTemporaryToken: () => createTemporaryToken,
186
185
  createTranscript: () => createTranscript,
187
186
  createTranscription: () => createTranscription,
188
187
  createVoiceRouter: () => createVoiceRouter,
@@ -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 {
@@ -5970,13 +5969,15 @@ var AssemblyAIAdapter = class extends BaseAdapter {
5970
5969
  const request = {
5971
5970
  ...options?.assemblyai,
5972
5971
  audio_url: audioUrl,
5972
+ // speech_models is required — default to universal-3-pro
5973
+ speech_models: options?.assemblyai?.speech_models ?? ["universal-3-pro"],
5973
5974
  // Enable punctuation and formatting by default
5974
5975
  punctuate: options?.assemblyai?.punctuate ?? true,
5975
5976
  format_text: options?.assemblyai?.format_text ?? true
5976
5977
  };
5977
5978
  if (options) {
5978
5979
  if (options.model) {
5979
- request.speech_model = options.model;
5980
+ request.speech_models = [options.model];
5980
5981
  }
5981
5982
  if (options.language) {
5982
5983
  const languageCode = options.language.includes("_") ? options.language : `${options.language}_us`;
@@ -5992,8 +5993,7 @@ var AssemblyAIAdapter = class extends BaseAdapter {
5992
5993
  }
5993
5994
  }
5994
5995
  if (options.customVocabulary && options.customVocabulary.length > 0) {
5995
- request.word_boost = options.customVocabulary;
5996
- request.boost_param = request.boost_param ?? "high";
5996
+ request.keyterms_prompt = options.customVocabulary;
5997
5997
  }
5998
5998
  if (options.summarization) {
5999
5999
  request.summarization = true;
@@ -11773,15 +11773,10 @@ var deepgramStreamingOnlyParams = import_zod2.z.object({
11773
11773
  // src/generated/assemblyai/api/assemblyAIAPI.zod.ts
11774
11774
  var assemblyAIAPI_zod_exports = {};
11775
11775
  __export(assemblyAIAPI_zod_exports, {
11776
- createTemporaryTokenBody: () => createTemporaryTokenBody,
11777
- createTemporaryTokenBodyExpiresInMin: () => createTemporaryTokenBodyExpiresInMin,
11778
- createTemporaryTokenResponse: () => createTemporaryTokenResponse,
11779
11776
  createTranscriptBody: () => createTranscriptBody,
11780
11777
  createTranscriptBodyAutoChaptersDefault: () => createTranscriptBodyAutoChaptersDefault,
11781
11778
  createTranscriptBodyAutoHighlightsDefault: () => createTranscriptBodyAutoHighlightsDefault,
11782
11779
  createTranscriptBodyContentSafetyConfidenceDefault: () => createTranscriptBodyContentSafetyConfidenceDefault,
11783
- createTranscriptBodyContentSafetyConfidenceMax: () => createTranscriptBodyContentSafetyConfidenceMax,
11784
- createTranscriptBodyContentSafetyConfidenceMin: () => createTranscriptBodyContentSafetyConfidenceMin,
11785
11780
  createTranscriptBodyContentSafetyDefault: () => createTranscriptBodyContentSafetyDefault,
11786
11781
  createTranscriptBodyCustomTopicsDefault: () => createTranscriptBodyCustomTopicsDefault,
11787
11782
  createTranscriptBodyDisfluenciesDefault: () => createTranscriptBodyDisfluenciesDefault,
@@ -11789,198 +11784,58 @@ __export(assemblyAIAPI_zod_exports, {
11789
11784
  createTranscriptBodyFilterProfanityDefault: () => createTranscriptBodyFilterProfanityDefault,
11790
11785
  createTranscriptBodyFormatTextDefault: () => createTranscriptBodyFormatTextDefault,
11791
11786
  createTranscriptBodyIabCategoriesDefault: () => createTranscriptBodyIabCategoriesDefault,
11792
- createTranscriptBodyLanguageCodeDefault: () => createTranscriptBodyLanguageCodeDefault,
11793
- createTranscriptBodyLanguageConfidenceThresholdDefault: () => createTranscriptBodyLanguageConfidenceThresholdDefault,
11794
- createTranscriptBodyLanguageConfidenceThresholdMax: () => createTranscriptBodyLanguageConfidenceThresholdMax,
11795
- createTranscriptBodyLanguageConfidenceThresholdMin: () => createTranscriptBodyLanguageConfidenceThresholdMin,
11796
11787
  createTranscriptBodyLanguageDetectionDefault: () => createTranscriptBodyLanguageDetectionDefault,
11788
+ createTranscriptBodyLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault: () => createTranscriptBodyLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault,
11789
+ createTranscriptBodyLanguageDetectionOptionsCodeSwitchingDefault: () => createTranscriptBodyLanguageDetectionOptionsCodeSwitchingDefault,
11790
+ createTranscriptBodyLanguageDetectionOptionsFallbackLanguageDefault: () => createTranscriptBodyLanguageDetectionOptionsFallbackLanguageDefault,
11797
11791
  createTranscriptBodyMultichannelDefault: () => createTranscriptBodyMultichannelDefault,
11798
11792
  createTranscriptBodyPunctuateDefault: () => createTranscriptBodyPunctuateDefault,
11799
11793
  createTranscriptBodyRedactPiiAudioDefault: () => createTranscriptBodyRedactPiiAudioDefault,
11794
+ createTranscriptBodyRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault: () => createTranscriptBodyRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault,
11800
11795
  createTranscriptBodyRedactPiiDefault: () => createTranscriptBodyRedactPiiDefault,
11801
- createTranscriptBodyRedactPiiSubDefault: () => createTranscriptBodyRedactPiiSubDefault,
11802
11796
  createTranscriptBodySentimentAnalysisDefault: () => createTranscriptBodySentimentAnalysisDefault,
11803
11797
  createTranscriptBodySpeakerLabelsDefault: () => createTranscriptBodySpeakerLabelsDefault,
11804
- createTranscriptBodySpeakersExpectedDefault: () => createTranscriptBodySpeakersExpectedDefault,
11805
- createTranscriptBodySpeechModelDefault: () => createTranscriptBodySpeechModelDefault,
11806
- createTranscriptBodySpeechThresholdDefault: () => createTranscriptBodySpeechThresholdDefault,
11807
- createTranscriptBodySpeechThresholdMax: () => createTranscriptBodySpeechThresholdMax,
11808
- createTranscriptBodySpeechThresholdMin: () => createTranscriptBodySpeechThresholdMin,
11798
+ createTranscriptBodySpeakerOptionsMinSpeakersExpectedDefault: () => createTranscriptBodySpeakerOptionsMinSpeakersExpectedDefault,
11799
+ createTranscriptBodySpeechUnderstandingRequestTranslationFormalDefault: () => createTranscriptBodySpeechUnderstandingRequestTranslationFormalDefault,
11800
+ createTranscriptBodySpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault: () => createTranscriptBodySpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault,
11809
11801
  createTranscriptBodySummarizationDefault: () => createTranscriptBodySummarizationDefault,
11810
- createTranscriptBodyWebhookAuthHeaderNameDefault: () => createTranscriptBodyWebhookAuthHeaderNameDefault,
11811
- createTranscriptBodyWebhookAuthHeaderValueDefault: () => createTranscriptBodyWebhookAuthHeaderValueDefault,
11802
+ createTranscriptBodyTemperatureDefault: () => createTranscriptBodyTemperatureDefault,
11812
11803
  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,
11804
+ createTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault: () => createTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault,
11805
+ createTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault: () => createTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault,
11806
+ createTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault: () => createTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault,
11807
+ createTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault: () => createTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault,
11808
+ createTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault: () => createTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault,
11809
+ createTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault: () => createTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault,
11848
11810
  deleteTranscriptParams: () => deleteTranscriptParams,
11849
11811
  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,
11812
+ deleteTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault: () => deleteTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault,
11813
+ deleteTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault: () => deleteTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault,
11814
+ deleteTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault: () => deleteTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault,
11815
+ deleteTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault: () => deleteTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault,
11816
+ deleteTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault: () => deleteTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault,
11817
+ deleteTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault: () => deleteTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault,
11891
11818
  getRedactedAudioParams: () => getRedactedAudioParams,
11892
11819
  getRedactedAudioResponse: () => getRedactedAudioResponse,
11893
11820
  getSubtitlesParams: () => getSubtitlesParams,
11894
11821
  getSubtitlesQueryParams: () => getSubtitlesQueryParams,
11822
+ getSubtitlesResponse: () => getSubtitlesResponse,
11895
11823
  getTranscriptParagraphsParams: () => getTranscriptParagraphsParams,
11896
11824
  getTranscriptParagraphsResponse: () => getTranscriptParagraphsResponse,
11897
- getTranscriptParagraphsResponseConfidenceMax: () => getTranscriptParagraphsResponseConfidenceMax,
11898
- getTranscriptParagraphsResponseConfidenceMin: () => getTranscriptParagraphsResponseConfidenceMin,
11899
- getTranscriptParagraphsResponseParagraphsItemConfidenceMax: () => getTranscriptParagraphsResponseParagraphsItemConfidenceMax,
11900
- getTranscriptParagraphsResponseParagraphsItemConfidenceMin: () => getTranscriptParagraphsResponseParagraphsItemConfidenceMin,
11901
- getTranscriptParagraphsResponseParagraphsItemWordsItemConfidenceMax: () => getTranscriptParagraphsResponseParagraphsItemWordsItemConfidenceMax,
11902
- getTranscriptParagraphsResponseParagraphsItemWordsItemConfidenceMin: () => getTranscriptParagraphsResponseParagraphsItemWordsItemConfidenceMin,
11903
11825
  getTranscriptParams: () => getTranscriptParams,
11904
11826
  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,
11827
+ getTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault: () => getTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault,
11828
+ getTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault: () => getTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault,
11829
+ getTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault: () => getTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault,
11830
+ getTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault: () => getTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault,
11831
+ getTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault: () => getTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault,
11832
+ getTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault: () => getTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault,
11940
11833
  getTranscriptSentencesParams: () => getTranscriptSentencesParams,
11941
11834
  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
11835
  listTranscriptsQueryLimitDefault: () => listTranscriptsQueryLimitDefault,
11976
- listTranscriptsQueryLimitMax: () => listTranscriptsQueryLimitMax,
11977
11836
  listTranscriptsQueryParams: () => listTranscriptsQueryParams,
11978
11837
  listTranscriptsQueryThrottledOnlyDefault: () => listTranscriptsQueryThrottledOnlyDefault,
11979
11838
  listTranscriptsResponse: () => listTranscriptsResponse,
11980
- listTranscriptsResponseTranscriptsItemCompletedRegExp: () => listTranscriptsResponseTranscriptsItemCompletedRegExp,
11981
- listTranscriptsResponseTranscriptsItemCreatedRegExp: () => listTranscriptsResponseTranscriptsItemCreatedRegExp,
11982
- purgeLemurRequestDataParams: () => purgeLemurRequestDataParams,
11983
- purgeLemurRequestDataResponse: () => purgeLemurRequestDataResponse,
11984
11839
  uploadFileResponse: () => uploadFileResponse,
11985
11840
  wordSearchParams: () => wordSearchParams,
11986
11841
  wordSearchQueryParams: () => wordSearchQueryParams,
@@ -11990,43 +11845,192 @@ var import_zod3 = require("zod");
11990
11845
  var uploadFileResponse = import_zod3.z.object({
11991
11846
  upload_url: import_zod3.z.string().describe("A URL that points to your audio file, accessible only by AssemblyAI's servers\n")
11992
11847
  });
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;
11848
+ var createTranscriptBodyAutoChaptersDefault = false;
12005
11849
  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
11850
  var createTranscriptBodyContentSafetyDefault = false;
12013
11851
  var createTranscriptBodyContentSafetyConfidenceDefault = 50;
12014
- var createTranscriptBodyContentSafetyConfidenceMin = 25;
12015
- var createTranscriptBodyContentSafetyConfidenceMax = 100;
11852
+ var createTranscriptBodyDisfluenciesDefault = false;
11853
+ var createTranscriptBodyEntityDetectionDefault = false;
11854
+ var createTranscriptBodyFilterProfanityDefault = false;
11855
+ var createTranscriptBodyFormatTextDefault = true;
12016
11856
  var createTranscriptBodyIabCategoriesDefault = false;
11857
+ var createTranscriptBodyLanguageDetectionDefault = false;
11858
+ var createTranscriptBodyLanguageDetectionOptionsFallbackLanguageDefault = "auto";
11859
+ var createTranscriptBodyLanguageDetectionOptionsCodeSwitchingDefault = false;
11860
+ var createTranscriptBodyLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault = 0.3;
11861
+ var createTranscriptBodyMultichannelDefault = false;
11862
+ var createTranscriptBodyPunctuateDefault = true;
11863
+ var createTranscriptBodyRedactPiiDefault = false;
11864
+ var createTranscriptBodyRedactPiiAudioDefault = false;
11865
+ var createTranscriptBodyRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault = false;
12017
11866
  var createTranscriptBodySentimentAnalysisDefault = false;
12018
- var createTranscriptBodyAutoChaptersDefault = false;
12019
- var createTranscriptBodyEntityDetectionDefault = false;
12020
- var createTranscriptBodySpeechThresholdDefault = 0;
12021
- var createTranscriptBodySpeechThresholdMin = 0;
12022
- var createTranscriptBodySpeechThresholdMax = 1;
11867
+ var createTranscriptBodySpeakerLabelsDefault = false;
11868
+ var createTranscriptBodySpeakerOptionsMinSpeakersExpectedDefault = 1;
11869
+ var createTranscriptBodySpeechUnderstandingRequestTranslationFormalDefault = true;
11870
+ var createTranscriptBodySpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault = false;
12023
11871
  var createTranscriptBodySummarizationDefault = false;
11872
+ var createTranscriptBodyTemperatureDefault = 0;
12024
11873
  var createTranscriptBodyCustomTopicsDefault = false;
12025
11874
  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([
11875
+ audio_end_at: import_zod3.z.number().optional().describe(
11876
+ "The point in time, in milliseconds, to stop transcribing in your media file. See [Set the start and end of the transcript](https://www.assemblyai.com/docs/pre-recorded-audio/set-the-start-and-end-of-the-transcript) for more details."
11877
+ ),
11878
+ audio_start_from: import_zod3.z.number().optional().describe(
11879
+ "The point in time, in milliseconds, to begin transcribing in your media file. See [Set the start and end of the transcript](https://www.assemblyai.com/docs/pre-recorded-audio/set-the-start-and-end-of-the-transcript) for more details."
11880
+ ),
11881
+ auto_chapters: import_zod3.z.boolean().optional().describe(
11882
+ "Enable [Auto Chapters](https://www.assemblyai.com/docs/speech-understanding/auto-chapters), can be true or false. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible chapter summaries. See the [updated Auto Chapters page](https://www.assemblyai.com/docs/speech-understanding/auto-chapters) for details.\n\nNote: This parameter is only supported for the Universal-2 model.\n"
11883
+ ),
11884
+ auto_highlights: import_zod3.z.boolean().optional().describe(
11885
+ "Enable [Key Phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases), either true or false"
11886
+ ),
11887
+ content_safety: import_zod3.z.boolean().optional().describe(
11888
+ "Enable [Content Moderation](https://www.assemblyai.com/docs/content-moderation), can be true or false"
11889
+ ),
11890
+ content_safety_confidence: import_zod3.z.number().default(createTranscriptBodyContentSafetyConfidenceDefault).describe(
11891
+ "The confidence threshold for the [Content Moderation](https://www.assemblyai.com/docs/content-moderation) model. Values must be between 25 and 100."
11892
+ ),
11893
+ custom_spelling: import_zod3.z.array(
11894
+ import_zod3.z.object({
11895
+ from: import_zod3.z.array(import_zod3.z.string()).describe("Words or phrases to replace"),
11896
+ to: import_zod3.z.string().describe("Word to replace with")
11897
+ }).describe(
11898
+ "Object containing words or phrases to replace, and the word or phrase to replace with"
11899
+ )
11900
+ ).optional().describe(
11901
+ "Customize how words are spelled and formatted using to and from values. See [Custom Spelling](https://www.assemblyai.com/docs/pre-recorded-audio/custom-spelling) for more details."
11902
+ ),
11903
+ disfluencies: import_zod3.z.boolean().optional().describe(
11904
+ 'Transcribe [Filler Words](https://www.assemblyai.com/docs/pre-recorded-audio/filler-words), like "umm", in your media file; can be true or false'
11905
+ ),
11906
+ domain: import_zod3.z.string().nullish().describe(
11907
+ 'Enable domain-specific transcription models to improve accuracy for specialized terminology. Set to `"medical-v1"` to enable [Medical Mode](https://www.assemblyai.com/docs/pre-recorded-audio/medical-mode) for improved accuracy of medical terms such as medications, procedures, conditions, and dosages.\n\nSupported languages: English (`en`), Spanish (`es`), German (`de`), French (`fr`). If used with an unsupported language, the parameter is ignored and a warning is returned.\n'
11908
+ ),
11909
+ entity_detection: import_zod3.z.boolean().optional().describe(
11910
+ "Enable [Entity Detection](https://www.assemblyai.com/docs/speech-understanding/entity-detection), can be true or false"
11911
+ ),
11912
+ filter_profanity: import_zod3.z.boolean().optional().describe(
11913
+ "Filter profanity from the transcribed text, can be true or false. See [Profanity Filtering](https://www.assemblyai.com/docs/profanity-filtering) for more details."
11914
+ ),
11915
+ format_text: import_zod3.z.boolean().default(createTranscriptBodyFormatTextDefault).describe(
11916
+ "Enable [Text Formatting](https://www.assemblyai.com/docs/pre-recorded-audio), can be true or false"
11917
+ ),
11918
+ iab_categories: import_zod3.z.boolean().optional().describe(
11919
+ "Enable [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection), can be true or false"
11920
+ ),
11921
+ keyterms_prompt: import_zod3.z.array(import_zod3.z.string()).optional().describe(
11922
+ "Improve accuracy with up to 200 (for Universal-2) or 1000 (for Universal-3 Pro) domain-specific words or phrases (maximum 6 words per phrase). See [Keyterms Prompting](https://www.assemblyai.com/docs/pre-recorded-audio/keyterms-prompting) for more details.\n"
11923
+ ),
11924
+ language_code: import_zod3.z.enum([
11925
+ "en",
11926
+ "en_au",
11927
+ "en_uk",
11928
+ "en_us",
11929
+ "es",
11930
+ "fr",
11931
+ "de",
11932
+ "it",
11933
+ "pt",
11934
+ "nl",
11935
+ "af",
11936
+ "sq",
11937
+ "am",
11938
+ "ar",
11939
+ "hy",
11940
+ "as",
11941
+ "az",
11942
+ "ba",
11943
+ "eu",
11944
+ "be",
11945
+ "bn",
11946
+ "bs",
11947
+ "br",
11948
+ "bg",
11949
+ "my",
11950
+ "ca",
11951
+ "zh",
11952
+ "hr",
11953
+ "cs",
11954
+ "da",
11955
+ "et",
11956
+ "fo",
11957
+ "fi",
11958
+ "gl",
11959
+ "ka",
11960
+ "el",
11961
+ "gu",
11962
+ "ht",
11963
+ "ha",
11964
+ "haw",
11965
+ "he",
11966
+ "hi",
11967
+ "hu",
11968
+ "is",
11969
+ "id",
11970
+ "ja",
11971
+ "jw",
11972
+ "kn",
11973
+ "kk",
11974
+ "km",
11975
+ "ko",
11976
+ "lo",
11977
+ "la",
11978
+ "lv",
11979
+ "ln",
11980
+ "lt",
11981
+ "lb",
11982
+ "mk",
11983
+ "mg",
11984
+ "ms",
11985
+ "ml",
11986
+ "mt",
11987
+ "mi",
11988
+ "mr",
11989
+ "mn",
11990
+ "ne",
11991
+ "no",
11992
+ "nn",
11993
+ "oc",
11994
+ "pa",
11995
+ "ps",
11996
+ "fa",
11997
+ "pl",
11998
+ "ro",
11999
+ "ru",
12000
+ "sa",
12001
+ "sr",
12002
+ "sn",
12003
+ "sd",
12004
+ "si",
12005
+ "sk",
12006
+ "sl",
12007
+ "so",
12008
+ "su",
12009
+ "sw",
12010
+ "sv",
12011
+ "tl",
12012
+ "tg",
12013
+ "ta",
12014
+ "tt",
12015
+ "te",
12016
+ "th",
12017
+ "bo",
12018
+ "tr",
12019
+ "tk",
12020
+ "uk",
12021
+ "ur",
12022
+ "uz",
12023
+ "vi",
12024
+ "cy",
12025
+ "yi",
12026
+ "yo"
12027
+ ]).describe(
12028
+ "The language of your audio file. Possible values are found in [Supported Languages](https://www.assemblyai.com/docs/pre-recorded-audio/supported-languages).\nThe default value is 'en_us'.\n"
12029
+ ).or(import_zod3.z.null()).optional().describe(
12030
+ "The language of your audio file. Possible values are found in [Supported Languages](https://www.assemblyai.com/docs/pre-recorded-audio/supported-languages).\nThe default value is 'en_us'.\n"
12031
+ ),
12032
+ language_codes: import_zod3.z.array(
12033
+ import_zod3.z.enum([
12030
12034
  "en",
12031
12035
  "en_au",
12032
12036
  "en_uk",
@@ -12130,53 +12134,365 @@ var createTranscriptBody = import_zod3.z.object({
12130
12134
  "yi",
12131
12135
  "yo"
12132
12136
  ]).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"
12137
+ "The language of your audio file. Possible values are found in [Supported Languages](https://www.assemblyai.com/docs/pre-recorded-audio/supported-languages).\nThe default value is 'en_us'.\n"
12138
+ )
12139
+ ).nullish().describe(
12140
+ "The language codes of your audio file. Used for [Code switching](/docs/speech-to-text/pre-recorded-audio/code-switching)\nOne of the values specified must be `en`.\n"
12141
+ ),
12142
+ language_confidence_threshold: import_zod3.z.number().optional().describe(
12143
+ "The confidence threshold for the automatically detected language.\nAn error will be returned if the language confidence is below this threshold.\nDefaults to 0. See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.\n"
12144
+ ),
12145
+ language_detection: import_zod3.z.boolean().optional().describe(
12146
+ "Enable [Automatic language detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection), either true or false."
12147
+ ),
12148
+ language_detection_options: import_zod3.z.object({
12149
+ expected_languages: import_zod3.z.array(import_zod3.z.string()).optional().describe(
12150
+ 'List of languages expected in the audio file. Defaults to `["all"]` when unspecified. See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.'
12156
12151
  ),
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"
12152
+ fallback_language: import_zod3.z.string().default(createTranscriptBodyLanguageDetectionOptionsFallbackLanguageDefault).describe(
12153
+ 'If the detected language of the audio file is not in the list of expected languages, the `fallback_language` is used. Specify `["auto"]` to let our model choose the fallback language from `expected_languages` with the highest confidence score. See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.\n'
12159
12154
  ),
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"
12155
+ code_switching: import_zod3.z.boolean().optional().describe(
12156
+ "Whether [code switching](/docs/speech-to-text/pre-recorded-audio/code-switching) should be detected.\n"
12162
12157
  ),
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"
12158
+ code_switching_confidence_threshold: import_zod3.z.number().default(
12159
+ createTranscriptBodyLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault
12160
+ ).describe(
12161
+ "The confidence threshold for [code switching](/docs/speech-to-text/pre-recorded-audio/code-switching) detection. If the code switching confidence is below this threshold, the transcript will be processed in the language with the highest `language_detection_confidence` score.\n"
12162
+ )
12163
+ }).optional().describe(
12164
+ "Specify options for [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection)."
12165
+ ),
12166
+ multichannel: import_zod3.z.boolean().optional().describe(
12167
+ "Enable [Multichannel](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) transcription, can be true or false."
12168
+ ),
12169
+ prompt: import_zod3.z.string().optional().describe(
12170
+ "Provide natural language prompting of up to 1,500 words of contextual information to the model. See the [Prompting Guide](https://www.assemblyai.com/docs/pre-recorded-audio/prompting) for best practices.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n"
12171
+ ),
12172
+ punctuate: import_zod3.z.boolean().default(createTranscriptBodyPunctuateDefault).describe(
12173
+ "Enable [Automatic Punctuation](https://www.assemblyai.com/docs/pre-recorded-audio), can be true or false"
12174
+ ),
12175
+ redact_pii: import_zod3.z.boolean().optional().describe(
12176
+ "Redact PII from the transcribed text using the Redact PII model, can be true or false. See [PII Redaction](https://www.assemblyai.com/docs/pii-redaction) for more details."
12177
+ ),
12178
+ redact_pii_audio: import_zod3.z.boolean().optional().describe(
12179
+ 'Generate a copy of the original media file with spoken PII "beeped" out, can be true or false. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more details.'
12180
+ ),
12181
+ redact_pii_audio_options: import_zod3.z.object({
12182
+ return_redacted_no_speech_audio: import_zod3.z.boolean().optional().describe(
12183
+ "By default, audio redaction provides redacted audio URLs only when speech is detected. However, if your use-case specifically requires redacted audio files even for silent audio files without any dialogue, you can opt to receive these URLs by setting this parameter to `true`."
12171
12184
  ),
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.'
12185
+ override_audio_redaction_method: import_zod3.z.enum(["silence"]).optional().describe(
12186
+ "Specify the method used to redact audio. By default, redacted audio uses a beep sound. Set to `silence` to replace PII with silence instead of a beep."
12187
+ )
12188
+ }).optional().describe(
12189
+ "Specify options for [PII redacted audio](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) files."
12190
+ ),
12191
+ redact_pii_audio_quality: import_zod3.z.enum(["mp3", "wav"]).optional().describe(
12192
+ "Controls the filetype of the audio created by redact_pii_audio. Currently supports mp3 (default) and wav. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more details."
12193
+ ),
12194
+ redact_pii_policies: import_zod3.z.array(
12195
+ import_zod3.z.enum([
12196
+ "account_number",
12197
+ "banking_information",
12198
+ "blood_type",
12199
+ "credit_card_cvv",
12200
+ "credit_card_expiration",
12201
+ "credit_card_number",
12202
+ "date",
12203
+ "date_interval",
12204
+ "date_of_birth",
12205
+ "drivers_license",
12206
+ "drug",
12207
+ "duration",
12208
+ "email_address",
12209
+ "event",
12210
+ "filename",
12211
+ "gender_sexuality",
12212
+ "healthcare_number",
12213
+ "injury",
12214
+ "ip_address",
12215
+ "language",
12216
+ "location",
12217
+ "marital_status",
12218
+ "medical_condition",
12219
+ "medical_process",
12220
+ "money_amount",
12221
+ "nationality",
12222
+ "number_sequence",
12223
+ "occupation",
12224
+ "organization",
12225
+ "passport_number",
12226
+ "password",
12227
+ "person_age",
12228
+ "person_name",
12229
+ "phone_number",
12230
+ "physical_attribute",
12231
+ "political_affiliation",
12232
+ "religion",
12233
+ "statistics",
12234
+ "time",
12235
+ "url",
12236
+ "us_social_security_number",
12237
+ "username",
12238
+ "vehicle_id",
12239
+ "zodiac_sign"
12240
+ ]).describe("The type of PII to redact")
12241
+ ).optional().describe(
12242
+ "The list of PII Redaction policies to enable. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction) for more details."
12243
+ ),
12244
+ redact_pii_sub: import_zod3.z.enum(["entity_name", "hash"]).describe(
12245
+ "The replacement logic for detected PII, can be `entity_name` or `hash`. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction) for more details."
12246
+ ).or(import_zod3.z.null()).optional().describe(
12247
+ "The replacement logic for detected PII, can be `entity_type` or `hash`. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction) for more details."
12248
+ ),
12249
+ sentiment_analysis: import_zod3.z.boolean().optional().describe(
12250
+ "Enable [Sentiment Analysis](https://www.assemblyai.com/docs/speech-understanding/sentiment-analysis), can be true or false"
12251
+ ),
12252
+ speaker_labels: import_zod3.z.boolean().optional().describe(
12253
+ "Enable [Speaker diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization), can be true or false"
12254
+ ),
12255
+ speaker_options: import_zod3.z.object({
12256
+ min_speakers_expected: import_zod3.z.number().default(createTranscriptBodySpeakerOptionsMinSpeakersExpectedDefault).describe(
12257
+ "The minimum number of speakers expected in the audio file. See [Set a range of possible speakers](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization#set-a-range-of-possible-speakers) for more details."
12174
12258
  ),
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."
12259
+ max_speakers_expected: import_zod3.z.number().optional().describe(
12260
+ "<Warning>Setting this parameter too high may hurt model accuracy</Warning>\nThe maximum number of speakers expected in the audio file. The default depends on audio duration: no limit for 0-2 minutes, 10 for 2-10 minutes, and 30 for 10+ minutes. See [Set a range of possible speakers](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization#set-a-range-of-possible-speakers) for more details.\n"
12261
+ )
12262
+ }).optional().describe(
12263
+ "Specify options for [Speaker diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization#set-a-range-of-possible-speakers). Use this to set a range of possible speakers."
12264
+ ),
12265
+ speakers_expected: import_zod3.z.number().nullish().describe(
12266
+ "Tells the speaker label model how many speakers it should attempt to identify. See [Set number of speakers expected](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization#set-number-of-speakers-expected) for more details."
12267
+ ),
12268
+ speech_models: import_zod3.z.array(
12269
+ import_zod3.z.string().describe(
12270
+ "The speech model to use for the transcription. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models."
12271
+ )
12272
+ ).describe(
12273
+ "List multiple speech models in priority order, allowing our system to automatically route your audio to the best available option. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models and routing behavior.\n"
12274
+ ),
12275
+ speech_threshold: import_zod3.z.number().nullish().describe(
12276
+ "Reject audio files that contain less than this fraction of speech.\nValid values are in the range [0, 1] inclusive. See [Speech Threshold](https://www.assemblyai.com/docs/speech-threshold) for more details.\n"
12277
+ ),
12278
+ speech_understanding: import_zod3.z.object({
12279
+ request: import_zod3.z.object({
12280
+ translation: import_zod3.z.object({
12281
+ target_languages: import_zod3.z.array(import_zod3.z.string()).describe(
12282
+ 'List of target language codes (e.g., `["es", "de"]`). See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for supported languages.'
12283
+ ),
12284
+ formal: import_zod3.z.boolean().default(createTranscriptBodySpeechUnderstandingRequestTranslationFormalDefault).describe(
12285
+ "Use formal language style. See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for more details."
12286
+ ),
12287
+ match_original_utterance: import_zod3.z.boolean().optional().describe(
12288
+ "When enabled with Speaker Labels, returns translated text in the utterances array. Each utterance will include a `translated_texts` key containing translations for each target language."
12289
+ )
12290
+ })
12291
+ }).describe(
12292
+ "Request body for [Translation](https://www.assemblyai.com/docs/speech-understanding/translation)."
12293
+ ).or(
12294
+ import_zod3.z.object({
12295
+ speaker_identification: import_zod3.z.object({
12296
+ speaker_type: import_zod3.z.enum(["role", "name"]).describe(
12297
+ "Type of speaker identification. See [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification) for details on each type."
12298
+ ),
12299
+ known_values: import_zod3.z.array(import_zod3.z.string()).optional().describe(
12300
+ 'Required if speaker_type is "role". Each value must be 35 characters or less.'
12301
+ ),
12302
+ speakers: import_zod3.z.array(
12303
+ import_zod3.z.object({
12304
+ role: import_zod3.z.string().optional().describe(
12305
+ 'The role of the speaker. Required when `speaker_type` is "role".'
12306
+ ),
12307
+ name: import_zod3.z.string().optional().describe(
12308
+ 'The name of the speaker. Required when `speaker_type` is "name".'
12309
+ ),
12310
+ description: import_zod3.z.string().optional().describe(
12311
+ "A description of the speaker to help the model identify them based on conversational context."
12312
+ )
12313
+ })
12314
+ ).optional().describe(
12315
+ "An array of speaker objects with metadata to improve identification accuracy. Each object should include a `role` or `name` (depending on `speaker_type`) and an optional `description` to help the model identify the speaker. You can also include any additional custom properties (e.g., `company`, `title`) to provide more context. Use this as an alternative to `known_values` when you want to provide additional context about each speaker."
12316
+ )
12317
+ })
12318
+ }).describe(
12319
+ "Request body for [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification)."
12320
+ )
12321
+ ).or(
12322
+ import_zod3.z.object({
12323
+ custom_formatting: import_zod3.z.object({
12324
+ date: import_zod3.z.string().optional().describe(
12325
+ 'Date format pattern (e.g., `"mm/dd/yyyy"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
12326
+ ),
12327
+ phone_number: import_zod3.z.string().optional().describe(
12328
+ 'Phone number format pattern (e.g., `"(xxx)xxx-xxxx"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
12329
+ ),
12330
+ email: import_zod3.z.string().optional().describe(
12331
+ 'Email format pattern (e.g., `"username@domain.com"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
12332
+ )
12333
+ })
12334
+ }).describe(
12335
+ "Request body for [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting)."
12336
+ )
12337
+ )
12338
+ }).optional().describe(
12339
+ "Enable speech understanding tasks like [Translation](https://www.assemblyai.com/docs/speech-understanding/translation), [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification), and [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting). See the task-specific docs for available options and configuration.\n"
12340
+ ),
12341
+ summarization: import_zod3.z.boolean().optional().describe(
12342
+ "Enable [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization), can be true or false. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details.\n\nNote: This parameter is only supported for the Universal-2 model.\n"
12343
+ ),
12344
+ summary_model: import_zod3.z.enum(["informative", "conversational", "catchy"]).optional().describe("The model to summarize the transcript"),
12345
+ summary_type: import_zod3.z.enum(["bullets", "bullets_verbose", "gist", "headline", "paragraph"]).optional().describe("The type of summary"),
12346
+ remove_audio_tags: import_zod3.z.enum(["all"]).describe(
12347
+ 'Remove [audio event tags](https://www.assemblyai.com/docs/pre-recorded-audio/universal-3-pro#audio-event-tags) from the transcript text. Set to `"all"` to remove all audio tags.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n'
12348
+ ).or(import_zod3.z.null()).optional().describe(
12349
+ 'Remove [audio event tags](https://www.assemblyai.com/docs/pre-recorded-audio/universal-3-pro#audio-event-tags) from the transcript text. Set to `"all"` to remove all audio tags.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n'
12350
+ ),
12351
+ temperature: import_zod3.z.number().optional().describe(
12352
+ "Control the amount of randomness injected into the model's response. See the [Prompting Guide](https://www.assemblyai.com/docs/pre-recorded-audio/prompting) for more details.\n\nNote: This parameter can only be used with the Universal-3 Pro model.\n"
12353
+ ),
12354
+ webhook_auth_header_name: import_zod3.z.string().nullish().describe(
12355
+ "The header name to be sent with the transcript completed or failed [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) requests"
12356
+ ),
12357
+ webhook_auth_header_value: import_zod3.z.string().nullish().describe(
12358
+ "The header value to send back with the transcript completed or failed [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) requests for added security"
12359
+ ),
12360
+ webhook_url: import_zod3.z.string().optional().describe(
12361
+ "The URL to which we send [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) requests.\n"
12362
+ ),
12363
+ custom_topics: import_zod3.z.boolean().optional().describe("This parameter does not currently have any functionality attached to it."),
12364
+ speech_model: import_zod3.z.string().describe(
12365
+ "The speech model to use for the transcription. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models."
12366
+ ).or(import_zod3.z.null()).optional().describe(
12367
+ "This parameter has been replaced with the `speech_models` parameter, learn more about the `speech_models` parameter [here](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model).\n"
12368
+ ),
12369
+ topics: import_zod3.z.array(import_zod3.z.string()).optional().describe("This parameter does not currently have any functionality attached to it."),
12370
+ audio_url: import_zod3.z.string().describe("The URL of the audio or video file to transcribe.")
12371
+ }).describe("The parameters for creating a transcript");
12372
+ var createTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault = "auto";
12373
+ var createTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault = false;
12374
+ var createTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault = 0.3;
12375
+ var createTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault = false;
12376
+ var createTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault = true;
12377
+ var createTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault = false;
12378
+ var createTranscriptResponse = import_zod3.z.object({
12379
+ audio_channels: import_zod3.z.number().optional().describe(
12380
+ "The number of audio channels in the audio file. This is only present when [multichannel](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) is enabled."
12381
+ ),
12382
+ audio_duration: import_zod3.z.number().nullish().describe("The duration of this transcript object's media file, in seconds"),
12383
+ audio_end_at: import_zod3.z.number().nullish().describe(
12384
+ "The point in time, in milliseconds, in the file at which the transcription was terminated. See [Set the start and end of the transcript](https://www.assemblyai.com/docs/pre-recorded-audio/set-the-start-and-end-of-the-transcript) for more details."
12385
+ ),
12386
+ audio_start_from: import_zod3.z.number().nullish().describe(
12387
+ "The point in time, in milliseconds, in the file at which the transcription was started. See [Set the start and end of the transcript](https://www.assemblyai.com/docs/pre-recorded-audio/set-the-start-and-end-of-the-transcript) for more details."
12388
+ ),
12389
+ audio_url: import_zod3.z.string().describe("The URL of the media that was transcribed"),
12390
+ auto_chapters: import_zod3.z.boolean().nullish().describe(
12391
+ "Whether [Auto Chapters](https://www.assemblyai.com/docs/speech-understanding/auto-chapters) is enabled, can be true or false. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible chapter summaries. See the [updated Auto Chapters page](https://www.assemblyai.com/docs/speech-understanding/auto-chapters) for details.\n\nNote: This parameter is only supported for the Universal-2 model.\n"
12392
+ ),
12393
+ auto_highlights: import_zod3.z.boolean().describe(
12394
+ "Whether [Key Phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases) is enabled, either true or false"
12395
+ ),
12396
+ auto_highlights_result: import_zod3.z.object({
12397
+ status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
12398
+ results: import_zod3.z.array(
12399
+ import_zod3.z.object({
12400
+ count: import_zod3.z.number().describe("The total number of times the key phrase appears in the audio file"),
12401
+ rank: import_zod3.z.number().describe(
12402
+ "The total relevancy to the overall audio file of this key phrase - a greater number means more relevant"
12403
+ ),
12404
+ text: import_zod3.z.string().describe("The text itself of the key phrase"),
12405
+ timestamps: import_zod3.z.array(
12406
+ import_zod3.z.object({
12407
+ start: import_zod3.z.number().describe("The start time in milliseconds"),
12408
+ end: import_zod3.z.number().describe("The end time in milliseconds")
12409
+ }).describe("Timestamp containing a start and end property in milliseconds")
12410
+ ).describe("The timestamp of the of the key phrase")
12411
+ })
12412
+ ).describe("A temporally-sequential array of Key Phrases")
12413
+ }).describe(
12414
+ "An array of results for the Key Phrases model, if it is enabled.\nSee [Key phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases) for more information.\n"
12415
+ ).or(import_zod3.z.null()).optional().describe(
12416
+ "An array of results for the Key Phrases model, if it is enabled.\nSee [Key Phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases) for more information.\n"
12417
+ ),
12418
+ chapters: import_zod3.z.array(
12419
+ import_zod3.z.object({
12420
+ gist: import_zod3.z.string().describe(
12421
+ "An ultra-short summary (just a few words) of the content spoken in the chapter"
12422
+ ),
12423
+ headline: import_zod3.z.string().describe("A single sentence summary of the content spoken during the chapter"),
12424
+ summary: import_zod3.z.string().describe("A one paragraph summary of the content spoken during the chapter"),
12425
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter"),
12426
+ end: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter")
12427
+ }).describe("Chapter of the audio file")
12428
+ ).nullish().describe(
12429
+ "An array of temporally sequential chapters for the audio file. See [Auto Chapters](https://www.assemblyai.com/docs/speech-understanding/auto-chapters) for more information."
12430
+ ),
12431
+ confidence: import_zod3.z.number().nullish().describe(
12432
+ "The confidence score for the transcript, between 0.0 (low confidence) and 1.0 (high confidence)"
12433
+ ),
12434
+ content_safety: import_zod3.z.boolean().nullish().describe(
12435
+ "Whether [Content Moderation](https://www.assemblyai.com/docs/content-moderation) is enabled, can be true or false"
12436
+ ),
12437
+ content_safety_labels: import_zod3.z.object({
12438
+ status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
12439
+ results: import_zod3.z.array(
12440
+ import_zod3.z.object({
12441
+ text: import_zod3.z.string().describe("The transcript of the section flagged by the Content Moderation model"),
12442
+ labels: import_zod3.z.array(
12443
+ import_zod3.z.object({
12444
+ label: import_zod3.z.string().describe("The label of the sensitive topic"),
12445
+ confidence: import_zod3.z.number().describe("The confidence score for the topic being discussed, from 0 to 1"),
12446
+ severity: import_zod3.z.number().describe("How severely the topic is discussed in the section, from 0 to 1")
12447
+ })
12448
+ ).describe(
12449
+ "An array of safety labels, one per sensitive topic that was detected in the section"
12450
+ ),
12451
+ sentences_idx_start: import_zod3.z.number().describe("The sentence index at which the section begins"),
12452
+ sentences_idx_end: import_zod3.z.number().describe("The sentence index at which the section ends"),
12453
+ timestamp: import_zod3.z.object({
12454
+ start: import_zod3.z.number().describe("The start time in milliseconds"),
12455
+ end: import_zod3.z.number().describe("The end time in milliseconds")
12456
+ }).describe("Timestamp containing a start and end property in milliseconds")
12457
+ })
12458
+ ).describe("An array of results for the Content Moderation model"),
12459
+ summary: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.number()).describe(
12460
+ "A summary of the Content Moderation confidence results for the entire audio file"
12177
12461
  ),
12178
- redact_pii_policies: import_zod3.z.array(
12179
- import_zod3.z.enum([
12462
+ severity_score_summary: import_zod3.z.record(
12463
+ import_zod3.z.string(),
12464
+ import_zod3.z.object({
12465
+ low: import_zod3.z.number(),
12466
+ medium: import_zod3.z.number(),
12467
+ high: import_zod3.z.number()
12468
+ })
12469
+ ).describe(
12470
+ "A summary of the Content Moderation severity results for the entire audio file"
12471
+ )
12472
+ }).describe(
12473
+ "An array of results for the Content Moderation model, if it is enabled.\nSee [Content moderation](https://www.assemblyai.com/docs/content-moderation) for more information.\n"
12474
+ ).or(import_zod3.z.null()).optional().describe(
12475
+ "An array of results for the Content Moderation model, if it is enabled.\nSee [Content moderation](https://www.assemblyai.com/docs/content-moderation) for more information.\n"
12476
+ ),
12477
+ custom_spelling: import_zod3.z.array(
12478
+ import_zod3.z.object({
12479
+ from: import_zod3.z.array(import_zod3.z.string()).describe("Words or phrases to replace"),
12480
+ to: import_zod3.z.string().describe("Word to replace with")
12481
+ }).describe(
12482
+ "Object containing words or phrases to replace, and the word or phrase to replace with"
12483
+ )
12484
+ ).nullish().describe(
12485
+ "Customize how words are spelled and formatted using to and from values. See [Custom Spelling](https://www.assemblyai.com/docs/pre-recorded-audio/custom-spelling) for more details."
12486
+ ),
12487
+ disfluencies: import_zod3.z.boolean().nullish().describe(
12488
+ 'Transcribe [Filler Words](https://www.assemblyai.com/docs/pre-recorded-audio/filler-words), like "umm", in your media file; can be true or false'
12489
+ ),
12490
+ domain: import_zod3.z.string().nullish().describe(
12491
+ 'The domain-specific model applied to the transcript. When set to `"medical-v1"`, [Medical Mode](https://www.assemblyai.com/docs/pre-recorded-audio/medical-mode) was used to improve accuracy for medical terminology.\n'
12492
+ ),
12493
+ entities: import_zod3.z.array(
12494
+ import_zod3.z.object({
12495
+ entity_type: import_zod3.z.enum([
12180
12496
  "account_number",
12181
12497
  "banking_information",
12182
12498
  "blood_type",
@@ -12221,103 +12537,59 @@ var createTranscriptBody = import_zod3.z.object({
12221
12537
  "username",
12222
12538
  "vehicle_id",
12223
12539
  "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"
12540
+ ]).describe("The type of entity for the detected entity"),
12541
+ text: import_zod3.z.string().describe("The text for the detected entity"),
12542
+ start: import_zod3.z.number().describe(
12543
+ "The starting time, in milliseconds, at which the detected entity appears in the audio file"
12544
+ ),
12545
+ end: import_zod3.z.number().describe(
12546
+ "The ending time, in milliseconds, for the detected entity in the audio file"
12254
12547
  )
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({
12548
+ }).describe("A detected entity")
12549
+ ).nullish().describe(
12550
+ "An array of results for the Entity Detection model, if it is enabled.\nSee [Entity detection](https://www.assemblyai.com/docs/speech-understanding/entity-detection) for more information.\n"
12551
+ ),
12552
+ entity_detection: import_zod3.z.boolean().nullish().describe(
12553
+ "Whether [Entity Detection](https://www.assemblyai.com/docs/speech-understanding/entity-detection) is enabled, can be true or false"
12554
+ ),
12555
+ error: import_zod3.z.string().optional().describe("Error message of why the transcript failed"),
12556
+ filter_profanity: import_zod3.z.boolean().nullish().describe(
12557
+ "Whether [Profanity Filtering](https://www.assemblyai.com/docs/profanity-filtering) is enabled, either true or false"
12558
+ ),
12559
+ format_text: import_zod3.z.boolean().nullish().describe(
12560
+ "Whether [Text Formatting](https://www.assemblyai.com/docs/pre-recorded-audio) is enabled, either true or false"
12561
+ ),
12562
+ iab_categories: import_zod3.z.boolean().nullish().describe(
12563
+ "Whether [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection) is enabled, can be true or false"
12564
+ ),
12565
+ iab_categories_result: import_zod3.z.object({
12566
+ status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
12567
+ results: import_zod3.z.array(
12568
+ import_zod3.z.object({
12569
+ text: import_zod3.z.string().describe("The text in the transcript in which a detected topic occurs"),
12570
+ labels: import_zod3.z.array(
12571
+ import_zod3.z.object({
12572
+ relevance: import_zod3.z.number().describe("How relevant the detected topic is of a detected topic"),
12573
+ label: import_zod3.z.string().describe(
12574
+ "The IAB taxonomical label for the label of the detected topic, where > denotes supertopic/subtopic relationship"
12575
+ )
12576
+ })
12577
+ ).optional().describe("An array of detected topics in the text"),
12578
+ timestamp: import_zod3.z.object({
12579
+ start: import_zod3.z.number().describe("The start time in milliseconds"),
12580
+ end: import_zod3.z.number().describe("The end time in milliseconds")
12581
+ }).optional().describe("Timestamp containing a start and end property in milliseconds")
12582
+ }).describe("The result of the topic detection model")
12583
+ ).describe("An array of results for the Topic Detection model"),
12584
+ summary: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.number()).describe("The overall relevance of topic to the entire audio file")
12585
+ }).describe(
12586
+ "The result of the Topic Detection model, if it is enabled.\nSee [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection) for more information.\n"
12587
+ ).or(import_zod3.z.null()).optional().describe(
12588
+ "The result of the Topic Detection model, if it is enabled.\nSee [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection) for more information.\n"
12589
+ ),
12317
12590
  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."
12591
+ keyterms_prompt: import_zod3.z.array(import_zod3.z.string()).optional().describe(
12592
+ "Improve accuracy with up to 200 (for Universal-2) or 1000 (for Universal-3 Pro) domain-specific words or phrases (maximum 6 words per phrase). See [Keyterms Prompting](https://www.assemblyai.com/docs/pre-recorded-audio/keyterms-prompting) for more details.\n"
12321
12593
  ),
12322
12594
  language_code: import_zod3.z.enum([
12323
12595
  "en",
@@ -12422,138 +12694,175 @@ var createTranscriptResponse = import_zod3.z.object({
12422
12694
  "cy",
12423
12695
  "yi",
12424
12696
  "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"
12697
+ ]).optional().describe(
12698
+ "The language of your audio file. Possible values are found in [Supported Languages](https://www.assemblyai.com/docs/pre-recorded-audio/supported-languages).\nThe default value is 'en_us'.\n"
12458
12699
  ),
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
- })
12700
+ language_codes: import_zod3.z.array(
12701
+ import_zod3.z.enum([
12702
+ "en",
12703
+ "en_au",
12704
+ "en_uk",
12705
+ "en_us",
12706
+ "es",
12707
+ "fr",
12708
+ "de",
12709
+ "it",
12710
+ "pt",
12711
+ "nl",
12712
+ "af",
12713
+ "sq",
12714
+ "am",
12715
+ "ar",
12716
+ "hy",
12717
+ "as",
12718
+ "az",
12719
+ "ba",
12720
+ "eu",
12721
+ "be",
12722
+ "bn",
12723
+ "bs",
12724
+ "br",
12725
+ "bg",
12726
+ "my",
12727
+ "ca",
12728
+ "zh",
12729
+ "hr",
12730
+ "cs",
12731
+ "da",
12732
+ "et",
12733
+ "fo",
12734
+ "fi",
12735
+ "gl",
12736
+ "ka",
12737
+ "el",
12738
+ "gu",
12739
+ "ht",
12740
+ "ha",
12741
+ "haw",
12742
+ "he",
12743
+ "hi",
12744
+ "hu",
12745
+ "is",
12746
+ "id",
12747
+ "ja",
12748
+ "jw",
12749
+ "kn",
12750
+ "kk",
12751
+ "km",
12752
+ "ko",
12753
+ "lo",
12754
+ "la",
12755
+ "lv",
12756
+ "ln",
12757
+ "lt",
12758
+ "lb",
12759
+ "mk",
12760
+ "mg",
12761
+ "ms",
12762
+ "ml",
12763
+ "mt",
12764
+ "mi",
12765
+ "mr",
12766
+ "mn",
12767
+ "ne",
12768
+ "no",
12769
+ "nn",
12770
+ "oc",
12771
+ "pa",
12772
+ "ps",
12773
+ "fa",
12774
+ "pl",
12775
+ "ro",
12776
+ "ru",
12777
+ "sa",
12778
+ "sr",
12779
+ "sn",
12780
+ "sd",
12781
+ "si",
12782
+ "sk",
12783
+ "sl",
12784
+ "so",
12785
+ "su",
12786
+ "sw",
12787
+ "sv",
12788
+ "tl",
12789
+ "tg",
12790
+ "ta",
12791
+ "tt",
12792
+ "te",
12793
+ "th",
12794
+ "bo",
12795
+ "tr",
12796
+ "tk",
12797
+ "uk",
12798
+ "ur",
12799
+ "uz",
12800
+ "vi",
12801
+ "cy",
12802
+ "yi",
12803
+ "yo"
12804
+ ]).describe(
12805
+ "The language of your audio file. Possible values are found in [Supported Languages](https://www.assemblyai.com/docs/pre-recorded-audio/supported-languages).\nThe default value is 'en_us'.\n"
12806
+ )
12486
12807
  ).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"
12808
+ "The language codes of your audio file. Used for [Code switching](/docs/speech-to-text/pre-recorded-audio/code-switching)\nOne of the values specified must be `en`.\n"
12488
12809
  ),
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)"
12810
+ language_confidence: import_zod3.z.number().nullable().describe(
12811
+ "The confidence score for the detected language, between 0.0 (low confidence) and 1.0 (high confidence). See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details."
12491
12812
  ),
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."
12813
+ language_confidence_threshold: import_zod3.z.number().nullable().describe(
12814
+ "The confidence threshold for the automatically detected language.\nAn error will be returned if the language confidence is below this threshold.\nSee [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.\n"
12501
12815
  ),
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"
12816
+ language_detection: import_zod3.z.boolean().nullish().describe(
12817
+ "Whether [Automatic language detection](/docs/pre-recorded-audio/automatic-language-detection) is enabled, either true or false"
12511
12818
  ),
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"
12819
+ language_detection_options: import_zod3.z.object({
12820
+ expected_languages: import_zod3.z.array(import_zod3.z.string()).optional().describe(
12821
+ 'List of languages expected in the audio file. Defaults to `["all"]` when unspecified. See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.'
12822
+ ),
12823
+ fallback_language: import_zod3.z.string().default(createTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault).describe(
12824
+ 'If the detected language of the audio file is not in the list of expected languages, the `fallback_language` is used. Specify `["auto"]` to let our model choose the fallback language from `expected_languages` with the highest confidence score. See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.\n'
12825
+ ),
12826
+ code_switching: import_zod3.z.boolean().optional().describe(
12827
+ "Whether [code switching](/docs/speech-to-text/pre-recorded-audio/code-switching) should be detected.\n"
12828
+ ),
12829
+ code_switching_confidence_threshold: import_zod3.z.number().default(
12830
+ createTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault
12831
+ ).describe(
12832
+ "The confidence threshold for [code switching](/docs/speech-to-text/pre-recorded-audio/code-switching) detection. If the code switching confidence is below this threshold, the transcript will be processed in the language with the highest `language_detection_confidence` score.\n"
12833
+ )
12834
+ }).optional().describe(
12835
+ "Specify options for [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection)."
12535
12836
  ),
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"
12837
+ multichannel: import_zod3.z.boolean().nullish().describe(
12838
+ "Whether [Multichannel transcription](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) was enabled in the transcription request, either true or false"
12538
12839
  ),
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"
12840
+ prompt: import_zod3.z.string().optional().describe(
12841
+ "Provide natural language prompting of up to 1,500 words of contextual information to the model. See the [Prompting Guide](https://www.assemblyai.com/docs/pre-recorded-audio/prompting) for best practices.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n"
12541
12842
  ),
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"
12843
+ punctuate: import_zod3.z.boolean().nullish().describe(
12844
+ "Whether [Automatic Punctuation](https://www.assemblyai.com/docs/pre-recorded-audio) is enabled, either true or false"
12546
12845
  ),
12547
12846
  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"
12847
+ "Whether [PII Redaction](https://www.assemblyai.com/docs/pii-redaction) is enabled, either true or false"
12549
12848
  ),
12550
12849
  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"
12850
+ "Whether a redacted version of the audio file was generated,\neither true or false. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more information.\n"
12851
+ ),
12852
+ redact_pii_audio_options: import_zod3.z.object({
12853
+ return_redacted_no_speech_audio: import_zod3.z.boolean().optional().describe(
12854
+ "By default, audio redaction provides redacted audio URLs only when speech is detected. However, if your use-case specifically requires redacted audio files even for silent audio files without any dialogue, you can opt to receive these URLs by setting this parameter to `true`."
12855
+ ),
12856
+ override_audio_redaction_method: import_zod3.z.enum(["silence"]).optional().describe(
12857
+ "Specify the method used to redact audio. By default, redacted audio uses a beep sound. Set to `silence` to replace PII with silence instead of a beep."
12858
+ )
12859
+ }).optional().describe(
12860
+ "The options for PII-redacted audio, if redact_pii_audio is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more information.\n"
12552
12861
  ),
12553
12862
  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."
12863
+ "Controls the filetype of the audio created by redact_pii_audio. Currently supports mp3 (default) and wav. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more details."
12555
12864
  ).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"
12865
+ "The audio quality of the PII-redacted audio file, if redact_pii_audio is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more information.\n"
12557
12866
  ),
12558
12867
  redact_pii_policies: import_zod3.z.array(
12559
12868
  import_zod3.z.enum([
@@ -12603,19 +12912,347 @@ var createTranscriptResponse = import_zod3.z.object({
12603
12912
  "zodiac_sign"
12604
12913
  ]).describe("The type of PII to redact")
12605
12914
  ).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"
12915
+ "The list of PII Redaction policies that were enabled, if PII Redaction is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/pii-redaction) for more information.\n"
12607
12916
  ),
12608
12917
  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.'
12918
+ "The replacement logic for detected PII, can be `entity_name` or `hash`. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction) for more details."
12919
+ ),
12920
+ sentiment_analysis: import_zod3.z.boolean().nullish().describe(
12921
+ "Whether [Sentiment Analysis](https://www.assemblyai.com/docs/speech-understanding/sentiment-analysis) is enabled, can be true or false"
12922
+ ),
12923
+ sentiment_analysis_results: import_zod3.z.array(
12924
+ import_zod3.z.object({
12925
+ text: import_zod3.z.string().describe("The transcript of the sentence"),
12926
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, of the sentence"),
12927
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, of the sentence"),
12928
+ sentiment: import_zod3.z.enum(["POSITIVE", "NEUTRAL", "NEGATIVE"]),
12929
+ confidence: import_zod3.z.number().describe(
12930
+ "The confidence score for the detected sentiment of the sentence, from 0 to 1"
12931
+ ),
12932
+ channel: import_zod3.z.string().nullish().describe(
12933
+ "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
12934
+ ),
12935
+ speaker: import_zod3.z.string().nullable().describe(
12936
+ "The speaker of the sentence if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
12937
+ )
12938
+ }).describe("The result of the Sentiment Analysis model")
12939
+ ).nullish().describe(
12940
+ "An array of results for the Sentiment Analysis model, if it is enabled.\nSee [Sentiment Analysis](https://www.assemblyai.com/docs/speech-understanding/sentiment-analysis) for more information.\n"
12610
12941
  ),
12611
12942
  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"
12943
+ "Whether [Speaker diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, can be true or false"
12613
12944
  ),
12614
12945
  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."
12946
+ "Tell the speaker label model how many speakers it should attempt to identify. See [Set number of speakers expected](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization#set-number-of-speakers-expected) for more details."
12947
+ ),
12948
+ speech_model_used: import_zod3.z.string().optional().describe(
12949
+ "The speech model to use for the transcription. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models."
12950
+ ),
12951
+ speech_models: import_zod3.z.array(
12952
+ import_zod3.z.string().describe(
12953
+ "The speech model to use for the transcription. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models."
12954
+ )
12955
+ ).nullish().describe(
12956
+ "List multiple speech models in priority order, allowing our system to automatically route your audio to the best available option. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models and routing behavior.\n"
12957
+ ),
12958
+ speech_threshold: import_zod3.z.number().nullish().describe(
12959
+ "Defaults to null. Reject audio files that contain less than this fraction of speech.\nValid values are in the range [0, 1] inclusive. See [Speech Threshold](https://www.assemblyai.com/docs/speech-threshold) for more details.\n"
12960
+ ),
12961
+ speech_understanding: import_zod3.z.object({
12962
+ request: import_zod3.z.object({
12963
+ translation: import_zod3.z.object({
12964
+ target_languages: import_zod3.z.array(import_zod3.z.string()).describe(
12965
+ 'List of target language codes (e.g., `["es", "de"]`). See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for supported languages.'
12966
+ ),
12967
+ formal: import_zod3.z.boolean().default(createTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault).describe(
12968
+ "Use formal language style. See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for more details."
12969
+ ),
12970
+ match_original_utterance: import_zod3.z.boolean().optional().describe(
12971
+ "When enabled with Speaker Labels, returns translated text in the utterances array. Each utterance will include a `translated_texts` key containing translations for each target language."
12972
+ )
12973
+ })
12974
+ }).describe(
12975
+ "Request body for [Translation](https://www.assemblyai.com/docs/speech-understanding/translation)."
12976
+ ).or(
12977
+ import_zod3.z.object({
12978
+ speaker_identification: import_zod3.z.object({
12979
+ speaker_type: import_zod3.z.enum(["role", "name"]).describe(
12980
+ "Type of speaker identification. See [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification) for details on each type."
12981
+ ),
12982
+ known_values: import_zod3.z.array(import_zod3.z.string()).optional().describe(
12983
+ 'Required if speaker_type is "role". Each value must be 35 characters or less.'
12984
+ ),
12985
+ speakers: import_zod3.z.array(
12986
+ import_zod3.z.object({
12987
+ role: import_zod3.z.string().optional().describe(
12988
+ 'The role of the speaker. Required when `speaker_type` is "role".'
12989
+ ),
12990
+ name: import_zod3.z.string().optional().describe(
12991
+ 'The name of the speaker. Required when `speaker_type` is "name".'
12992
+ ),
12993
+ description: import_zod3.z.string().optional().describe(
12994
+ "A description of the speaker to help the model identify them based on conversational context."
12995
+ )
12996
+ })
12997
+ ).optional().describe(
12998
+ "An array of speaker objects with metadata to improve identification accuracy. Each object should include a `role` or `name` (depending on `speaker_type`) and an optional `description` to help the model identify the speaker. You can also include any additional custom properties (e.g., `company`, `title`) to provide more context. Use this as an alternative to `known_values` when you want to provide additional context about each speaker."
12999
+ )
13000
+ })
13001
+ }).describe(
13002
+ "Request body for [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification)."
13003
+ )
13004
+ ).or(
13005
+ import_zod3.z.object({
13006
+ custom_formatting: import_zod3.z.object({
13007
+ date: import_zod3.z.string().optional().describe(
13008
+ 'Date format pattern (e.g., `"mm/dd/yyyy"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
13009
+ ),
13010
+ phone_number: import_zod3.z.string().optional().describe(
13011
+ 'Phone number format pattern (e.g., `"(xxx)xxx-xxxx"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
13012
+ ),
13013
+ email: import_zod3.z.string().optional().describe(
13014
+ 'Email format pattern (e.g., `"username@domain.com"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
13015
+ )
13016
+ })
13017
+ }).describe(
13018
+ "Request body for [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting)."
13019
+ )
13020
+ ).optional(),
13021
+ response: import_zod3.z.object({
13022
+ translation: import_zod3.z.object({
13023
+ status: import_zod3.z.string().optional()
13024
+ }).optional()
13025
+ }).or(
13026
+ import_zod3.z.object({
13027
+ speaker_identification: import_zod3.z.object({
13028
+ mapping: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional().describe(
13029
+ 'A mapping of the original generic speaker labels (e.g., "A", "B") to the identified speaker names or roles.'
13030
+ ),
13031
+ status: import_zod3.z.string().optional()
13032
+ }).optional()
13033
+ })
13034
+ ).or(
13035
+ import_zod3.z.object({
13036
+ custom_formatting: import_zod3.z.object({
13037
+ mapping: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional(),
13038
+ formatted_text: import_zod3.z.string().optional()
13039
+ }).optional()
13040
+ })
13041
+ ).optional()
13042
+ }).optional().describe(
13043
+ "Speech understanding tasks like [Translation](https://www.assemblyai.com/docs/speech-understanding/translation), [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification), and [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting). See the task-specific docs for available options and configuration.\n"
13044
+ ),
13045
+ status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).describe(
13046
+ "The status of your transcript. Possible values are queued, processing, completed, or error."
13047
+ ),
13048
+ summarization: import_zod3.z.boolean().describe(
13049
+ "Whether [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization) is enabled, either true or false. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details.\n\nNote: This parameter is only supported for the Universal-2 model.\n"
13050
+ ),
13051
+ summary: import_zod3.z.string().nullish().describe(
13052
+ "The generated summary of the media file, if [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization) is enabled. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details."
13053
+ ),
13054
+ summary_model: import_zod3.z.string().nullish().describe(
13055
+ "The Summarization model used to generate the summary,\nif [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization#summary-models) is enabled. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details.\n"
13056
+ ),
13057
+ summary_type: import_zod3.z.string().nullish().describe(
13058
+ "The type of summary generated, if [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization#summary-types) is enabled. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details."
13059
+ ),
13060
+ remove_audio_tags: import_zod3.z.enum(["all"]).describe(
13061
+ "Whether [audio event tags](https://www.assemblyai.com/docs/pre-recorded-audio/universal-3-pro#audio-event-tags) were removed from the transcript text.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n"
13062
+ ).or(import_zod3.z.null()).optional().describe(
13063
+ "Whether [audio event tags](https://www.assemblyai.com/docs/pre-recorded-audio/universal-3-pro#audio-event-tags) were removed from the transcript text.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n"
13064
+ ),
13065
+ temperature: import_zod3.z.number().nullish().describe(
13066
+ "The temperature that was used for the model's response. See the [Prompting Guide](https://www.assemblyai.com/docs/pre-recorded-audio/prompting) for more details.\n\nNote: This parameter can only be used with the Universal-3 Pro model.\n"
13067
+ ),
13068
+ text: import_zod3.z.string().nullish().describe("The textual transcript of your media file"),
13069
+ throttled: import_zod3.z.boolean().nullish().describe(
13070
+ "True while a request is throttled and false when a request is no longer throttled"
13071
+ ),
13072
+ utterances: import_zod3.z.array(
13073
+ import_zod3.z.object({
13074
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this utterance"),
13075
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, of the utterance in the audio file"),
13076
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, of the utterance in the audio file"),
13077
+ text: import_zod3.z.string().describe("The text for this utterance"),
13078
+ words: import_zod3.z.array(
13079
+ import_zod3.z.object({
13080
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
13081
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
13082
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
13083
+ text: import_zod3.z.string().describe("The text of the word"),
13084
+ channel: import_zod3.z.string().nullish().describe(
13085
+ "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13086
+ ),
13087
+ speaker: import_zod3.z.string().nullable().describe(
13088
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
13089
+ )
13090
+ })
13091
+ ).describe("The words in the utterance."),
13092
+ channel: import_zod3.z.string().nullish().describe(
13093
+ "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13094
+ ),
13095
+ speaker: import_zod3.z.string().describe(
13096
+ 'The speaker of this utterance, where each speaker is assigned a sequential capital letter - e.g. "A" for Speaker A, "B" for Speaker B, etc.'
13097
+ ),
13098
+ translated_texts: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional().describe(
13099
+ 'Translations keyed by language code (e.g., `{"es": "Texto traducido", "de": "\xDCbersetzter Text"}`). Only present when `match_original_utterance` is enabled with translation.'
13100
+ )
13101
+ })
13102
+ ).nullish().describe(
13103
+ "When multichannel or speaker_labels is enabled, a list of turn-by-turn utterance objects.\nSee [Speaker diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) and [Multichannel transcription](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) for more information.\n"
13104
+ ),
13105
+ webhook_auth: import_zod3.z.boolean().describe(
13106
+ "Whether [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) authentication details were provided"
13107
+ ),
13108
+ webhook_auth_header_name: import_zod3.z.string().nullish().describe(
13109
+ "The header name to be sent with the transcript completed or failed [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) requests"
13110
+ ),
13111
+ webhook_status_code: import_zod3.z.number().nullish().describe(
13112
+ "The status code we received from your server when delivering the transcript completed or failed [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) request, if a webhook URL was provided"
13113
+ ),
13114
+ webhook_url: import_zod3.z.string().nullish().describe(
13115
+ "The URL to which we send [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) requests.\n"
13116
+ ),
13117
+ words: import_zod3.z.array(
13118
+ import_zod3.z.object({
13119
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
13120
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
13121
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
13122
+ text: import_zod3.z.string().describe("The text of the word"),
13123
+ channel: import_zod3.z.string().nullish().describe(
13124
+ "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13125
+ ),
13126
+ speaker: import_zod3.z.string().nullable().describe(
13127
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
13128
+ )
13129
+ })
13130
+ ).nullish().describe(
13131
+ "An array of temporally-sequential word objects, one for each word in the transcript.\n"
13132
+ ),
13133
+ acoustic_model: import_zod3.z.string().describe("This parameter does not currently have any functionality attached to it."),
13134
+ custom_topics: import_zod3.z.boolean().nullish().describe("This parameter does not currently have any functionality attached to it."),
13135
+ language_model: import_zod3.z.string().describe("This parameter does not currently have any functionality attached to it."),
13136
+ speech_model: import_zod3.z.string().describe(
13137
+ "The speech model to use for the transcription. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models."
13138
+ ).or(import_zod3.z.null()).describe(
13139
+ "This parameter has been replaced with the `speech_models` parameter, learn more about the `speech_models` parameter [here](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model).\n"
13140
+ ),
13141
+ speed_boost: import_zod3.z.boolean().nullish().describe("This parameter does not currently have any functionality attached to it."),
13142
+ topics: import_zod3.z.array(import_zod3.z.string()).optional().describe("This parameter does not currently have any functionality attached to it."),
13143
+ translated_texts: import_zod3.z.object({
13144
+ language_code: import_zod3.z.string().optional().describe("Translated text for this language code")
13145
+ }).optional().describe(
13146
+ "Translated text keyed by language code. See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for more details."
13147
+ )
13148
+ }).describe("A transcript object");
13149
+ var listTranscriptsQueryLimitDefault = 10;
13150
+ var listTranscriptsQueryThrottledOnlyDefault = false;
13151
+ var listTranscriptsQueryParams = import_zod3.z.object({
13152
+ limit: import_zod3.z.number().default(listTranscriptsQueryLimitDefault).describe("Maximum amount of transcripts to retrieve"),
13153
+ status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).optional().describe("Filter by transcript status"),
13154
+ created_on: import_zod3.z.string().date().optional().describe("Only get transcripts created on this date"),
13155
+ before_id: import_zod3.z.string().uuid().optional().describe("Get transcripts that were created before this transcript ID"),
13156
+ after_id: import_zod3.z.string().uuid().optional().describe("Get transcripts that were created after this transcript ID"),
13157
+ throttled_only: import_zod3.z.boolean().optional().describe("Only get throttled transcripts, overrides the status filter")
13158
+ });
13159
+ var listTranscriptsResponse = import_zod3.z.object({
13160
+ page_details: import_zod3.z.object({
13161
+ limit: import_zod3.z.number().describe("The number of results this page is limited to"),
13162
+ result_count: import_zod3.z.number().describe("The actual number of results in the page"),
13163
+ current_url: import_zod3.z.string().describe("The URL used to retrieve the current page of transcripts"),
13164
+ prev_url: import_zod3.z.string().nullable().describe(
13165
+ "The URL to the next page of transcripts. The previous URL always points to a page with older transcripts."
13166
+ ),
13167
+ next_url: import_zod3.z.string().nullable().describe(
13168
+ "The URL to the next page of transcripts. The next URL always points to a page with newer transcripts."
13169
+ )
13170
+ }).describe(
13171
+ "Details of the transcript page. Transcripts are sorted from newest to oldest. The previous URL always points to a page with older transcripts."
13172
+ ),
13173
+ transcripts: import_zod3.z.array(
13174
+ import_zod3.z.object({
13175
+ id: import_zod3.z.string().uuid().describe("The unique identifier for the transcript"),
13176
+ resource_url: import_zod3.z.string().describe("The URL to retrieve the transcript"),
13177
+ status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).describe(
13178
+ "The status of your transcript. Possible values are queued, processing, completed, or error."
13179
+ ),
13180
+ created: import_zod3.z.string().datetime({}).describe("The date and time the transcript was created"),
13181
+ completed: import_zod3.z.string().datetime({}).optional().describe("The date and time the transcript was completed"),
13182
+ audio_url: import_zod3.z.string().describe("The URL to the audio file"),
13183
+ error: import_zod3.z.string().nullable().describe("Error message of why the transcript failed")
13184
+ })
13185
+ ).describe("An array of transcripts")
13186
+ }).describe(
13187
+ "A list of transcripts. Transcripts are sorted from newest to oldest. The previous URL always points to a page with older transcripts."
13188
+ );
13189
+ var getTranscriptParams = import_zod3.z.object({
13190
+ transcript_id: import_zod3.z.string().describe("ID of the transcript")
13191
+ });
13192
+ var getTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault = "auto";
13193
+ var getTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault = false;
13194
+ var getTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault = 0.3;
13195
+ var getTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault = false;
13196
+ var getTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault = true;
13197
+ var getTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault = false;
13198
+ var getTranscriptResponse = import_zod3.z.object({
13199
+ audio_channels: import_zod3.z.number().optional().describe(
13200
+ "The number of audio channels in the audio file. This is only present when [multichannel](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) is enabled."
13201
+ ),
13202
+ audio_duration: import_zod3.z.number().nullish().describe("The duration of this transcript object's media file, in seconds"),
13203
+ audio_end_at: import_zod3.z.number().nullish().describe(
13204
+ "The point in time, in milliseconds, in the file at which the transcription was terminated. See [Set the start and end of the transcript](https://www.assemblyai.com/docs/pre-recorded-audio/set-the-start-and-end-of-the-transcript) for more details."
13205
+ ),
13206
+ audio_start_from: import_zod3.z.number().nullish().describe(
13207
+ "The point in time, in milliseconds, in the file at which the transcription was started. See [Set the start and end of the transcript](https://www.assemblyai.com/docs/pre-recorded-audio/set-the-start-and-end-of-the-transcript) for more details."
13208
+ ),
13209
+ audio_url: import_zod3.z.string().describe("The URL of the media that was transcribed"),
13210
+ auto_chapters: import_zod3.z.boolean().nullish().describe(
13211
+ "Whether [Auto Chapters](https://www.assemblyai.com/docs/speech-understanding/auto-chapters) is enabled, can be true or false. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible chapter summaries. See the [updated Auto Chapters page](https://www.assemblyai.com/docs/speech-understanding/auto-chapters) for details.\n\nNote: This parameter is only supported for the Universal-2 model.\n"
13212
+ ),
13213
+ auto_highlights: import_zod3.z.boolean().describe(
13214
+ "Whether [Key Phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases) is enabled, either true or false"
13215
+ ),
13216
+ auto_highlights_result: import_zod3.z.object({
13217
+ status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
13218
+ results: import_zod3.z.array(
13219
+ import_zod3.z.object({
13220
+ count: import_zod3.z.number().describe("The total number of times the key phrase appears in the audio file"),
13221
+ rank: import_zod3.z.number().describe(
13222
+ "The total relevancy to the overall audio file of this key phrase - a greater number means more relevant"
13223
+ ),
13224
+ text: import_zod3.z.string().describe("The text itself of the key phrase"),
13225
+ timestamps: import_zod3.z.array(
13226
+ import_zod3.z.object({
13227
+ start: import_zod3.z.number().describe("The start time in milliseconds"),
13228
+ end: import_zod3.z.number().describe("The end time in milliseconds")
13229
+ }).describe("Timestamp containing a start and end property in milliseconds")
13230
+ ).describe("The timestamp of the of the key phrase")
13231
+ })
13232
+ ).describe("A temporally-sequential array of Key Phrases")
13233
+ }).describe(
13234
+ "An array of results for the Key Phrases model, if it is enabled.\nSee [Key phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases) for more information.\n"
13235
+ ).or(import_zod3.z.null()).optional().describe(
13236
+ "An array of results for the Key Phrases model, if it is enabled.\nSee [Key Phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases) for more information.\n"
13237
+ ),
13238
+ chapters: import_zod3.z.array(
13239
+ import_zod3.z.object({
13240
+ gist: import_zod3.z.string().describe(
13241
+ "An ultra-short summary (just a few words) of the content spoken in the chapter"
13242
+ ),
13243
+ headline: import_zod3.z.string().describe("A single sentence summary of the content spoken during the chapter"),
13244
+ summary: import_zod3.z.string().describe("A one paragraph summary of the content spoken during the chapter"),
13245
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter"),
13246
+ end: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter")
13247
+ }).describe("Chapter of the audio file")
13248
+ ).nullish().describe(
13249
+ "An array of temporally sequential chapters for the audio file. See [Auto Chapters](https://www.assemblyai.com/docs/speech-understanding/auto-chapters) for more information."
13250
+ ),
13251
+ confidence: import_zod3.z.number().nullish().describe(
13252
+ "The confidence score for the transcript, between 0.0 (low confidence) and 1.0 (high confidence)"
12616
13253
  ),
12617
13254
  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"
13255
+ "Whether [Content Moderation](https://www.assemblyai.com/docs/content-moderation) is enabled, can be true or false"
12619
13256
  ),
12620
13257
  content_safety_labels: import_zod3.z.object({
12621
13258
  status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
@@ -12625,16 +13262,8 @@ var createTranscriptResponse = import_zod3.z.object({
12625
13262
  labels: import_zod3.z.array(
12626
13263
  import_zod3.z.object({
12627
13264
  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")
13265
+ confidence: import_zod3.z.number().describe("The confidence score for the topic being discussed, from 0 to 1"),
13266
+ severity: import_zod3.z.number().describe("How severely the topic is discussed in the section, from 0 to 1")
12638
13267
  })
12639
13268
  ).describe(
12640
13269
  "An array of safety labels, one per sensitive topic that was detected in the section"
@@ -12647,128 +13276,39 @@ var createTranscriptResponse = import_zod3.z.object({
12647
13276
  }).describe("Timestamp containing a start and end property in milliseconds")
12648
13277
  })
12649
13278
  ).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(
13279
+ summary: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.number()).describe(
12656
13280
  "A summary of the Content Moderation confidence results for the entire audio file"
12657
13281
  ),
12658
13282
  severity_score_summary: import_zod3.z.record(
12659
13283
  import_zod3.z.string(),
12660
13284
  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)
13285
+ low: import_zod3.z.number(),
13286
+ medium: import_zod3.z.number(),
13287
+ high: import_zod3.z.number()
12664
13288
  })
12665
13289
  ).describe(
12666
13290
  "A summary of the Content Moderation severity results for the entire audio file"
12667
13291
  )
12668
13292
  }).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"
13293
+ "An array of results for the Content Moderation model, if it is enabled.\nSee [Content moderation](https://www.assemblyai.com/docs/content-moderation) for more information.\n"
12670
13294
  ).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"
13295
+ "An array of results for the Content Moderation model, if it is enabled.\nSee [Content moderation](https://www.assemblyai.com/docs/content-moderation) for more information.\n"
12707
13296
  ),
12708
13297
  custom_spelling: import_zod3.z.array(
12709
13298
  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"),
13299
+ from: import_zod3.z.array(import_zod3.z.string()).describe("Words or phrases to replace"),
12711
13300
  to: import_zod3.z.string().describe("Word to replace with")
12712
13301
  }).describe(
12713
13302
  "Object containing words or phrases to replace, and the word or phrase to replace with"
12714
13303
  )
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
13304
  ).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"
13305
+ "Customize how words are spelled and formatted using to and from values. See [Custom Spelling](https://www.assemblyai.com/docs/pre-recorded-audio/custom-spelling) for more details."
12769
13306
  ),
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"
13307
+ disfluencies: import_zod3.z.boolean().nullish().describe(
13308
+ 'Transcribe [Filler Words](https://www.assemblyai.com/docs/pre-recorded-audio/filler-words), like "umm", in your media file; can be true or false'
13309
+ ),
13310
+ domain: import_zod3.z.string().nullish().describe(
13311
+ 'The domain-specific model applied to the transcript. When set to `"medical-v1"`, [Medical Mode](https://www.assemblyai.com/docs/pre-recorded-audio/medical-mode) was used to improve accuracy for medical terminology.\n'
12772
13312
  ),
12773
13313
  entities: import_zod3.z.array(
12774
13314
  import_zod3.z.object({
@@ -12827,108 +13367,49 @@ var createTranscriptResponse = import_zod3.z.object({
12827
13367
  )
12828
13368
  }).describe("A detected entity")
12829
13369
  ).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"
13370
+ "An array of results for the Entity Detection model, if it is enabled.\nSee [Entity detection](https://www.assemblyai.com/docs/speech-understanding/entity-detection) for more information.\n"
12834
13371
  ),
12835
- throttled: import_zod3.z.boolean().nullish().describe(
12836
- "True while a request is throttled and false when a request is no longer throttled"
13372
+ entity_detection: import_zod3.z.boolean().nullish().describe(
13373
+ "Whether [Entity Detection](https://www.assemblyai.com/docs/speech-understanding/entity-detection) is enabled, can be true or false"
12837
13374
  ),
12838
13375
  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
- )
13376
+ filter_profanity: import_zod3.z.boolean().nullish().describe(
13377
+ "Whether [Profanity Filtering](https://www.assemblyai.com/docs/profanity-filtering) is enabled, either true or false"
13378
+ ),
13379
+ format_text: import_zod3.z.boolean().nullish().describe(
13380
+ "Whether [Text Formatting](https://www.assemblyai.com/docs/pre-recorded-audio) is enabled, either true or false"
13381
+ ),
13382
+ iab_categories: import_zod3.z.boolean().nullish().describe(
13383
+ "Whether [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection) is enabled, can be true or false"
13384
+ ),
13385
+ iab_categories_result: import_zod3.z.object({
13386
+ status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
13387
+ results: import_zod3.z.array(
13388
+ import_zod3.z.object({
13389
+ text: import_zod3.z.string().describe("The text in the transcript in which a detected topic occurs"),
13390
+ labels: import_zod3.z.array(
13391
+ import_zod3.z.object({
13392
+ relevance: import_zod3.z.number().describe("How relevant the detected topic is of a detected topic"),
13393
+ label: import_zod3.z.string().describe(
13394
+ "The IAB taxonomical label for the label of the detected topic, where > denotes supertopic/subtopic relationship"
13395
+ )
13396
+ })
13397
+ ).optional().describe("An array of detected topics in the text"),
13398
+ timestamp: import_zod3.z.object({
13399
+ start: import_zod3.z.number().describe("The start time in milliseconds"),
13400
+ end: import_zod3.z.number().describe("The end time in milliseconds")
13401
+ }).optional().describe("Timestamp containing a start and end property in milliseconds")
13402
+ }).describe("The result of the topic detection model")
13403
+ ).describe("An array of results for the Topic Detection model"),
13404
+ summary: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.number()).describe("The overall relevance of topic to the entire audio file")
12870
13405
  }).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."
13406
+ "The result of the Topic Detection model, if it is enabled.\nSee [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection) for more information.\n"
13407
+ ).or(import_zod3.z.null()).optional().describe(
13408
+ "The result of the Topic Detection model, if it is enabled.\nSee [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection) for more information.\n"
12872
13409
  ),
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
13410
  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."
13411
+ keyterms_prompt: import_zod3.z.array(import_zod3.z.string()).optional().describe(
13412
+ "Improve accuracy with up to 200 (for Universal-2) or 1000 (for Universal-3 Pro) domain-specific words or phrases (maximum 6 words per phrase). See [Keyterms Prompting](https://www.assemblyai.com/docs/pre-recorded-audio/keyterms-prompting) for more details.\n"
12932
13413
  ),
12933
13414
  language_code: import_zod3.z.enum([
12934
13415
  "en",
@@ -13033,138 +13514,175 @@ var getTranscriptResponse = import_zod3.z.object({
13033
13514
  "cy",
13034
13515
  "yi",
13035
13516
  "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"
13517
+ ]).optional().describe(
13518
+ "The language of your audio file. Possible values are found in [Supported Languages](https://www.assemblyai.com/docs/pre-recorded-audio/supported-languages).\nThe default value is 'en_us'.\n"
13069
13519
  ),
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
- })
13520
+ language_codes: import_zod3.z.array(
13521
+ import_zod3.z.enum([
13522
+ "en",
13523
+ "en_au",
13524
+ "en_uk",
13525
+ "en_us",
13526
+ "es",
13527
+ "fr",
13528
+ "de",
13529
+ "it",
13530
+ "pt",
13531
+ "nl",
13532
+ "af",
13533
+ "sq",
13534
+ "am",
13535
+ "ar",
13536
+ "hy",
13537
+ "as",
13538
+ "az",
13539
+ "ba",
13540
+ "eu",
13541
+ "be",
13542
+ "bn",
13543
+ "bs",
13544
+ "br",
13545
+ "bg",
13546
+ "my",
13547
+ "ca",
13548
+ "zh",
13549
+ "hr",
13550
+ "cs",
13551
+ "da",
13552
+ "et",
13553
+ "fo",
13554
+ "fi",
13555
+ "gl",
13556
+ "ka",
13557
+ "el",
13558
+ "gu",
13559
+ "ht",
13560
+ "ha",
13561
+ "haw",
13562
+ "he",
13563
+ "hi",
13564
+ "hu",
13565
+ "is",
13566
+ "id",
13567
+ "ja",
13568
+ "jw",
13569
+ "kn",
13570
+ "kk",
13571
+ "km",
13572
+ "ko",
13573
+ "lo",
13574
+ "la",
13575
+ "lv",
13576
+ "ln",
13577
+ "lt",
13578
+ "lb",
13579
+ "mk",
13580
+ "mg",
13581
+ "ms",
13582
+ "ml",
13583
+ "mt",
13584
+ "mi",
13585
+ "mr",
13586
+ "mn",
13587
+ "ne",
13588
+ "no",
13589
+ "nn",
13590
+ "oc",
13591
+ "pa",
13592
+ "ps",
13593
+ "fa",
13594
+ "pl",
13595
+ "ro",
13596
+ "ru",
13597
+ "sa",
13598
+ "sr",
13599
+ "sn",
13600
+ "sd",
13601
+ "si",
13602
+ "sk",
13603
+ "sl",
13604
+ "so",
13605
+ "su",
13606
+ "sw",
13607
+ "sv",
13608
+ "tl",
13609
+ "tg",
13610
+ "ta",
13611
+ "tt",
13612
+ "te",
13613
+ "th",
13614
+ "bo",
13615
+ "tr",
13616
+ "tk",
13617
+ "uk",
13618
+ "ur",
13619
+ "uz",
13620
+ "vi",
13621
+ "cy",
13622
+ "yi",
13623
+ "yo"
13624
+ ]).describe(
13625
+ "The language of your audio file. Possible values are found in [Supported Languages](https://www.assemblyai.com/docs/pre-recorded-audio/supported-languages).\nThe default value is 'en_us'.\n"
13626
+ )
13097
13627
  ).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)"
13628
+ "The language codes of your audio file. Used for [Code switching](/docs/speech-to-text/pre-recorded-audio/code-switching)\nOne of the values specified must be `en`.\n"
13102
13629
  ),
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."
13630
+ language_confidence: import_zod3.z.number().nullable().describe(
13631
+ "The confidence score for the detected language, between 0.0 (low confidence) and 1.0 (high confidence). See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details."
13112
13632
  ),
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"
13633
+ language_confidence_threshold: import_zod3.z.number().nullable().describe(
13634
+ "The confidence threshold for the automatically detected language.\nAn error will be returned if the language confidence is below this threshold.\nSee [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.\n"
13118
13635
  ),
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"
13636
+ language_detection: import_zod3.z.boolean().nullish().describe(
13637
+ "Whether [Automatic language detection](/docs/pre-recorded-audio/automatic-language-detection) is enabled, either true or false"
13122
13638
  ),
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"
13639
+ language_detection_options: import_zod3.z.object({
13640
+ expected_languages: import_zod3.z.array(import_zod3.z.string()).optional().describe(
13641
+ 'List of languages expected in the audio file. Defaults to `["all"]` when unspecified. See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.'
13642
+ ),
13643
+ fallback_language: import_zod3.z.string().default(getTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault).describe(
13644
+ 'If the detected language of the audio file is not in the list of expected languages, the `fallback_language` is used. Specify `["auto"]` to let our model choose the fallback language from `expected_languages` with the highest confidence score. See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.\n'
13645
+ ),
13646
+ code_switching: import_zod3.z.boolean().optional().describe(
13647
+ "Whether [code switching](/docs/speech-to-text/pre-recorded-audio/code-switching) should be detected.\n"
13648
+ ),
13649
+ code_switching_confidence_threshold: import_zod3.z.number().default(
13650
+ getTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault
13651
+ ).describe(
13652
+ "The confidence threshold for [code switching](/docs/speech-to-text/pre-recorded-audio/code-switching) detection. If the code switching confidence is below this threshold, the transcript will be processed in the language with the highest `language_detection_confidence` score.\n"
13653
+ )
13654
+ }).optional().describe(
13655
+ "Specify options for [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection)."
13146
13656
  ),
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"
13657
+ multichannel: import_zod3.z.boolean().nullish().describe(
13658
+ "Whether [Multichannel transcription](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) was enabled in the transcription request, either true or false"
13149
13659
  ),
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"
13660
+ prompt: import_zod3.z.string().optional().describe(
13661
+ "Provide natural language prompting of up to 1,500 words of contextual information to the model. See the [Prompting Guide](https://www.assemblyai.com/docs/pre-recorded-audio/prompting) for best practices.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n"
13152
13662
  ),
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"
13663
+ punctuate: import_zod3.z.boolean().nullish().describe(
13664
+ "Whether [Automatic Punctuation](https://www.assemblyai.com/docs/pre-recorded-audio) is enabled, either true or false"
13157
13665
  ),
13158
13666
  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"
13667
+ "Whether [PII Redaction](https://www.assemblyai.com/docs/pii-redaction) is enabled, either true or false"
13160
13668
  ),
13161
13669
  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"
13670
+ "Whether a redacted version of the audio file was generated,\neither true or false. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more information.\n"
13671
+ ),
13672
+ redact_pii_audio_options: import_zod3.z.object({
13673
+ return_redacted_no_speech_audio: import_zod3.z.boolean().optional().describe(
13674
+ "By default, audio redaction provides redacted audio URLs only when speech is detected. However, if your use-case specifically requires redacted audio files even for silent audio files without any dialogue, you can opt to receive these URLs by setting this parameter to `true`."
13675
+ ),
13676
+ override_audio_redaction_method: import_zod3.z.enum(["silence"]).optional().describe(
13677
+ "Specify the method used to redact audio. By default, redacted audio uses a beep sound. Set to `silence` to replace PII with silence instead of a beep."
13678
+ )
13679
+ }).optional().describe(
13680
+ "The options for PII-redacted audio, if redact_pii_audio is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more information.\n"
13163
13681
  ),
13164
13682
  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."
13683
+ "Controls the filetype of the audio created by redact_pii_audio. Currently supports mp3 (default) and wav. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more details."
13166
13684
  ).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"
13685
+ "The audio quality of the PII-redacted audio file, if redact_pii_audio is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more information.\n"
13168
13686
  ),
13169
13687
  redact_pii_policies: import_zod3.z.array(
13170
13688
  import_zod3.z.enum([
@@ -13214,19 +13732,307 @@ var getTranscriptResponse = import_zod3.z.object({
13214
13732
  "zodiac_sign"
13215
13733
  ]).describe("The type of PII to redact")
13216
13734
  ).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"
13735
+ "The list of PII Redaction policies that were enabled, if PII Redaction is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/pii-redaction) for more information.\n"
13218
13736
  ),
13219
13737
  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.'
13738
+ "The replacement logic for detected PII, can be `entity_name` or `hash`. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction) for more details."
13739
+ ),
13740
+ sentiment_analysis: import_zod3.z.boolean().nullish().describe(
13741
+ "Whether [Sentiment Analysis](https://www.assemblyai.com/docs/speech-understanding/sentiment-analysis) is enabled, can be true or false"
13742
+ ),
13743
+ sentiment_analysis_results: import_zod3.z.array(
13744
+ import_zod3.z.object({
13745
+ text: import_zod3.z.string().describe("The transcript of the sentence"),
13746
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, of the sentence"),
13747
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, of the sentence"),
13748
+ sentiment: import_zod3.z.enum(["POSITIVE", "NEUTRAL", "NEGATIVE"]),
13749
+ confidence: import_zod3.z.number().describe(
13750
+ "The confidence score for the detected sentiment of the sentence, from 0 to 1"
13751
+ ),
13752
+ channel: import_zod3.z.string().nullish().describe(
13753
+ "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13754
+ ),
13755
+ speaker: import_zod3.z.string().nullable().describe(
13756
+ "The speaker of the sentence if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
13757
+ )
13758
+ }).describe("The result of the Sentiment Analysis model")
13759
+ ).nullish().describe(
13760
+ "An array of results for the Sentiment Analysis model, if it is enabled.\nSee [Sentiment Analysis](https://www.assemblyai.com/docs/speech-understanding/sentiment-analysis) for more information.\n"
13221
13761
  ),
13222
13762
  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"
13763
+ "Whether [Speaker diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, can be true or false"
13224
13764
  ),
13225
13765
  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."
13766
+ "Tell the speaker label model how many speakers it should attempt to identify. See [Set number of speakers expected](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization#set-number-of-speakers-expected) for more details."
13767
+ ),
13768
+ speech_model_used: import_zod3.z.string().optional().describe(
13769
+ "The speech model to use for the transcription. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models."
13770
+ ),
13771
+ speech_models: import_zod3.z.array(
13772
+ import_zod3.z.string().describe(
13773
+ "The speech model to use for the transcription. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models."
13774
+ )
13775
+ ).nullish().describe(
13776
+ "List multiple speech models in priority order, allowing our system to automatically route your audio to the best available option. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models and routing behavior.\n"
13777
+ ),
13778
+ speech_threshold: import_zod3.z.number().nullish().describe(
13779
+ "Defaults to null. Reject audio files that contain less than this fraction of speech.\nValid values are in the range [0, 1] inclusive. See [Speech Threshold](https://www.assemblyai.com/docs/speech-threshold) for more details.\n"
13780
+ ),
13781
+ speech_understanding: import_zod3.z.object({
13782
+ request: import_zod3.z.object({
13783
+ translation: import_zod3.z.object({
13784
+ target_languages: import_zod3.z.array(import_zod3.z.string()).describe(
13785
+ 'List of target language codes (e.g., `["es", "de"]`). See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for supported languages.'
13786
+ ),
13787
+ formal: import_zod3.z.boolean().default(getTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault).describe(
13788
+ "Use formal language style. See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for more details."
13789
+ ),
13790
+ match_original_utterance: import_zod3.z.boolean().optional().describe(
13791
+ "When enabled with Speaker Labels, returns translated text in the utterances array. Each utterance will include a `translated_texts` key containing translations for each target language."
13792
+ )
13793
+ })
13794
+ }).describe(
13795
+ "Request body for [Translation](https://www.assemblyai.com/docs/speech-understanding/translation)."
13796
+ ).or(
13797
+ import_zod3.z.object({
13798
+ speaker_identification: import_zod3.z.object({
13799
+ speaker_type: import_zod3.z.enum(["role", "name"]).describe(
13800
+ "Type of speaker identification. See [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification) for details on each type."
13801
+ ),
13802
+ known_values: import_zod3.z.array(import_zod3.z.string()).optional().describe(
13803
+ 'Required if speaker_type is "role". Each value must be 35 characters or less.'
13804
+ ),
13805
+ speakers: import_zod3.z.array(
13806
+ import_zod3.z.object({
13807
+ role: import_zod3.z.string().optional().describe(
13808
+ 'The role of the speaker. Required when `speaker_type` is "role".'
13809
+ ),
13810
+ name: import_zod3.z.string().optional().describe(
13811
+ 'The name of the speaker. Required when `speaker_type` is "name".'
13812
+ ),
13813
+ description: import_zod3.z.string().optional().describe(
13814
+ "A description of the speaker to help the model identify them based on conversational context."
13815
+ )
13816
+ })
13817
+ ).optional().describe(
13818
+ "An array of speaker objects with metadata to improve identification accuracy. Each object should include a `role` or `name` (depending on `speaker_type`) and an optional `description` to help the model identify the speaker. You can also include any additional custom properties (e.g., `company`, `title`) to provide more context. Use this as an alternative to `known_values` when you want to provide additional context about each speaker."
13819
+ )
13820
+ })
13821
+ }).describe(
13822
+ "Request body for [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification)."
13823
+ )
13824
+ ).or(
13825
+ import_zod3.z.object({
13826
+ custom_formatting: import_zod3.z.object({
13827
+ date: import_zod3.z.string().optional().describe(
13828
+ 'Date format pattern (e.g., `"mm/dd/yyyy"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
13829
+ ),
13830
+ phone_number: import_zod3.z.string().optional().describe(
13831
+ 'Phone number format pattern (e.g., `"(xxx)xxx-xxxx"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
13832
+ ),
13833
+ email: import_zod3.z.string().optional().describe(
13834
+ 'Email format pattern (e.g., `"username@domain.com"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
13835
+ )
13836
+ })
13837
+ }).describe(
13838
+ "Request body for [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting)."
13839
+ )
13840
+ ).optional(),
13841
+ response: import_zod3.z.object({
13842
+ translation: import_zod3.z.object({
13843
+ status: import_zod3.z.string().optional()
13844
+ }).optional()
13845
+ }).or(
13846
+ import_zod3.z.object({
13847
+ speaker_identification: import_zod3.z.object({
13848
+ mapping: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional().describe(
13849
+ 'A mapping of the original generic speaker labels (e.g., "A", "B") to the identified speaker names or roles.'
13850
+ ),
13851
+ status: import_zod3.z.string().optional()
13852
+ }).optional()
13853
+ })
13854
+ ).or(
13855
+ import_zod3.z.object({
13856
+ custom_formatting: import_zod3.z.object({
13857
+ mapping: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional(),
13858
+ formatted_text: import_zod3.z.string().optional()
13859
+ }).optional()
13860
+ })
13861
+ ).optional()
13862
+ }).optional().describe(
13863
+ "Speech understanding tasks like [Translation](https://www.assemblyai.com/docs/speech-understanding/translation), [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification), and [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting). See the task-specific docs for available options and configuration.\n"
13864
+ ),
13865
+ status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).describe(
13866
+ "The status of your transcript. Possible values are queued, processing, completed, or error."
13867
+ ),
13868
+ summarization: import_zod3.z.boolean().describe(
13869
+ "Whether [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization) is enabled, either true or false. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details.\n\nNote: This parameter is only supported for the Universal-2 model.\n"
13870
+ ),
13871
+ summary: import_zod3.z.string().nullish().describe(
13872
+ "The generated summary of the media file, if [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization) is enabled. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details."
13873
+ ),
13874
+ summary_model: import_zod3.z.string().nullish().describe(
13875
+ "The Summarization model used to generate the summary,\nif [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization#summary-models) is enabled. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details.\n"
13876
+ ),
13877
+ summary_type: import_zod3.z.string().nullish().describe(
13878
+ "The type of summary generated, if [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization#summary-types) is enabled. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details."
13879
+ ),
13880
+ remove_audio_tags: import_zod3.z.enum(["all"]).describe(
13881
+ "Whether [audio event tags](https://www.assemblyai.com/docs/pre-recorded-audio/universal-3-pro#audio-event-tags) were removed from the transcript text.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n"
13882
+ ).or(import_zod3.z.null()).optional().describe(
13883
+ "Whether [audio event tags](https://www.assemblyai.com/docs/pre-recorded-audio/universal-3-pro#audio-event-tags) were removed from the transcript text.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n"
13884
+ ),
13885
+ temperature: import_zod3.z.number().nullish().describe(
13886
+ "The temperature that was used for the model's response. See the [Prompting Guide](https://www.assemblyai.com/docs/pre-recorded-audio/prompting) for more details.\n\nNote: This parameter can only be used with the Universal-3 Pro model.\n"
13887
+ ),
13888
+ text: import_zod3.z.string().nullish().describe("The textual transcript of your media file"),
13889
+ throttled: import_zod3.z.boolean().nullish().describe(
13890
+ "True while a request is throttled and false when a request is no longer throttled"
13891
+ ),
13892
+ utterances: import_zod3.z.array(
13893
+ import_zod3.z.object({
13894
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this utterance"),
13895
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, of the utterance in the audio file"),
13896
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, of the utterance in the audio file"),
13897
+ text: import_zod3.z.string().describe("The text for this utterance"),
13898
+ words: import_zod3.z.array(
13899
+ import_zod3.z.object({
13900
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
13901
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
13902
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
13903
+ text: import_zod3.z.string().describe("The text of the word"),
13904
+ channel: import_zod3.z.string().nullish().describe(
13905
+ "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13906
+ ),
13907
+ speaker: import_zod3.z.string().nullable().describe(
13908
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
13909
+ )
13910
+ })
13911
+ ).describe("The words in the utterance."),
13912
+ channel: import_zod3.z.string().nullish().describe(
13913
+ "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13914
+ ),
13915
+ speaker: import_zod3.z.string().describe(
13916
+ 'The speaker of this utterance, where each speaker is assigned a sequential capital letter - e.g. "A" for Speaker A, "B" for Speaker B, etc.'
13917
+ ),
13918
+ translated_texts: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional().describe(
13919
+ 'Translations keyed by language code (e.g., `{"es": "Texto traducido", "de": "\xDCbersetzter Text"}`). Only present when `match_original_utterance` is enabled with translation.'
13920
+ )
13921
+ })
13922
+ ).nullish().describe(
13923
+ "When multichannel or speaker_labels is enabled, a list of turn-by-turn utterance objects.\nSee [Speaker diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) and [Multichannel transcription](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) for more information.\n"
13924
+ ),
13925
+ webhook_auth: import_zod3.z.boolean().describe(
13926
+ "Whether [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) authentication details were provided"
13927
+ ),
13928
+ webhook_auth_header_name: import_zod3.z.string().nullish().describe(
13929
+ "The header name to be sent with the transcript completed or failed [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) requests"
13930
+ ),
13931
+ webhook_status_code: import_zod3.z.number().nullish().describe(
13932
+ "The status code we received from your server when delivering the transcript completed or failed [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) request, if a webhook URL was provided"
13933
+ ),
13934
+ webhook_url: import_zod3.z.string().nullish().describe(
13935
+ "The URL to which we send [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) requests.\n"
13936
+ ),
13937
+ words: import_zod3.z.array(
13938
+ import_zod3.z.object({
13939
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
13940
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
13941
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
13942
+ text: import_zod3.z.string().describe("The text of the word"),
13943
+ channel: import_zod3.z.string().nullish().describe(
13944
+ "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13945
+ ),
13946
+ speaker: import_zod3.z.string().nullable().describe(
13947
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
13948
+ )
13949
+ })
13950
+ ).nullish().describe(
13951
+ "An array of temporally-sequential word objects, one for each word in the transcript.\n"
13952
+ ),
13953
+ acoustic_model: import_zod3.z.string().describe("This parameter does not currently have any functionality attached to it."),
13954
+ custom_topics: import_zod3.z.boolean().nullish().describe("This parameter does not currently have any functionality attached to it."),
13955
+ language_model: import_zod3.z.string().describe("This parameter does not currently have any functionality attached to it."),
13956
+ speech_model: import_zod3.z.string().describe(
13957
+ "The speech model to use for the transcription. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models."
13958
+ ).or(import_zod3.z.null()).describe(
13959
+ "This parameter has been replaced with the `speech_models` parameter, learn more about the `speech_models` parameter [here](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model).\n"
13960
+ ),
13961
+ speed_boost: import_zod3.z.boolean().nullish().describe("This parameter does not currently have any functionality attached to it."),
13962
+ topics: import_zod3.z.array(import_zod3.z.string()).optional().describe("This parameter does not currently have any functionality attached to it."),
13963
+ translated_texts: import_zod3.z.object({
13964
+ language_code: import_zod3.z.string().optional().describe("Translated text for this language code")
13965
+ }).optional().describe(
13966
+ "Translated text keyed by language code. See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for more details."
13967
+ )
13968
+ }).describe("A transcript object");
13969
+ var deleteTranscriptParams = import_zod3.z.object({
13970
+ transcript_id: import_zod3.z.string().describe("ID of the transcript")
13971
+ });
13972
+ var deleteTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault = "auto";
13973
+ var deleteTranscriptResponseLanguageDetectionOptionsCodeSwitchingDefault = false;
13974
+ var deleteTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault = 0.3;
13975
+ var deleteTranscriptResponseRedactPiiAudioOptionsReturnRedactedNoSpeechAudioDefault = false;
13976
+ var deleteTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault = true;
13977
+ var deleteTranscriptResponseSpeechUnderstandingRequestTranslationMatchOriginalUtteranceDefault = false;
13978
+ var deleteTranscriptResponse = import_zod3.z.object({
13979
+ audio_channels: import_zod3.z.number().optional().describe(
13980
+ "The number of audio channels in the audio file. This is only present when [multichannel](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) is enabled."
13981
+ ),
13982
+ audio_duration: import_zod3.z.number().nullish().describe("The duration of this transcript object's media file, in seconds"),
13983
+ audio_end_at: import_zod3.z.number().nullish().describe(
13984
+ "The point in time, in milliseconds, in the file at which the transcription was terminated. See [Set the start and end of the transcript](https://www.assemblyai.com/docs/pre-recorded-audio/set-the-start-and-end-of-the-transcript) for more details."
13985
+ ),
13986
+ audio_start_from: import_zod3.z.number().nullish().describe(
13987
+ "The point in time, in milliseconds, in the file at which the transcription was started. See [Set the start and end of the transcript](https://www.assemblyai.com/docs/pre-recorded-audio/set-the-start-and-end-of-the-transcript) for more details."
13988
+ ),
13989
+ audio_url: import_zod3.z.string().describe("The URL of the media that was transcribed"),
13990
+ auto_chapters: import_zod3.z.boolean().nullish().describe(
13991
+ "Whether [Auto Chapters](https://www.assemblyai.com/docs/speech-understanding/auto-chapters) is enabled, can be true or false. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible chapter summaries. See the [updated Auto Chapters page](https://www.assemblyai.com/docs/speech-understanding/auto-chapters) for details.\n\nNote: This parameter is only supported for the Universal-2 model.\n"
13992
+ ),
13993
+ auto_highlights: import_zod3.z.boolean().describe(
13994
+ "Whether [Key Phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases) is enabled, either true or false"
13995
+ ),
13996
+ auto_highlights_result: import_zod3.z.object({
13997
+ status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
13998
+ results: import_zod3.z.array(
13999
+ import_zod3.z.object({
14000
+ count: import_zod3.z.number().describe("The total number of times the key phrase appears in the audio file"),
14001
+ rank: import_zod3.z.number().describe(
14002
+ "The total relevancy to the overall audio file of this key phrase - a greater number means more relevant"
14003
+ ),
14004
+ text: import_zod3.z.string().describe("The text itself of the key phrase"),
14005
+ timestamps: import_zod3.z.array(
14006
+ import_zod3.z.object({
14007
+ start: import_zod3.z.number().describe("The start time in milliseconds"),
14008
+ end: import_zod3.z.number().describe("The end time in milliseconds")
14009
+ }).describe("Timestamp containing a start and end property in milliseconds")
14010
+ ).describe("The timestamp of the of the key phrase")
14011
+ })
14012
+ ).describe("A temporally-sequential array of Key Phrases")
14013
+ }).describe(
14014
+ "An array of results for the Key Phrases model, if it is enabled.\nSee [Key phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases) for more information.\n"
14015
+ ).or(import_zod3.z.null()).optional().describe(
14016
+ "An array of results for the Key Phrases model, if it is enabled.\nSee [Key Phrases](https://www.assemblyai.com/docs/speech-understanding/key-phrases) for more information.\n"
14017
+ ),
14018
+ chapters: import_zod3.z.array(
14019
+ import_zod3.z.object({
14020
+ gist: import_zod3.z.string().describe(
14021
+ "An ultra-short summary (just a few words) of the content spoken in the chapter"
14022
+ ),
14023
+ headline: import_zod3.z.string().describe("A single sentence summary of the content spoken during the chapter"),
14024
+ summary: import_zod3.z.string().describe("A one paragraph summary of the content spoken during the chapter"),
14025
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter"),
14026
+ end: import_zod3.z.number().describe("The starting time, in milliseconds, for the chapter")
14027
+ }).describe("Chapter of the audio file")
14028
+ ).nullish().describe(
14029
+ "An array of temporally sequential chapters for the audio file. See [Auto Chapters](https://www.assemblyai.com/docs/speech-understanding/auto-chapters) for more information."
14030
+ ),
14031
+ confidence: import_zod3.z.number().nullish().describe(
14032
+ "The confidence score for the transcript, between 0.0 (low confidence) and 1.0 (high confidence)"
13227
14033
  ),
13228
14034
  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"
14035
+ "Whether [Content Moderation](https://www.assemblyai.com/docs/content-moderation) is enabled, can be true or false"
13230
14036
  ),
13231
14037
  content_safety_labels: import_zod3.z.object({
13232
14038
  status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
@@ -13236,12 +14042,8 @@ var getTranscriptResponse = import_zod3.z.object({
13236
14042
  labels: import_zod3.z.array(
13237
14043
  import_zod3.z.object({
13238
14044
  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")
14045
+ confidence: import_zod3.z.number().describe("The confidence score for the topic being discussed, from 0 to 1"),
14046
+ severity: import_zod3.z.number().describe("How severely the topic is discussed in the section, from 0 to 1")
13245
14047
  })
13246
14048
  ).describe(
13247
14049
  "An array of safety labels, one per sensitive topic that was detected in the section"
@@ -13254,128 +14056,39 @@ var getTranscriptResponse = import_zod3.z.object({
13254
14056
  }).describe("Timestamp containing a start and end property in milliseconds")
13255
14057
  })
13256
14058
  ).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(
14059
+ summary: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.number()).describe(
13263
14060
  "A summary of the Content Moderation confidence results for the entire audio file"
13264
14061
  ),
13265
14062
  severity_score_summary: import_zod3.z.record(
13266
14063
  import_zod3.z.string(),
13267
14064
  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)
14065
+ low: import_zod3.z.number(),
14066
+ medium: import_zod3.z.number(),
14067
+ high: import_zod3.z.number()
13271
14068
  })
13272
14069
  ).describe(
13273
14070
  "A summary of the Content Moderation severity results for the entire audio file"
13274
14071
  )
13275
14072
  }).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"
14073
+ "An array of results for the Content Moderation model, if it is enabled.\nSee [Content moderation](https://www.assemblyai.com/docs/content-moderation) for more information.\n"
13312
14074
  ).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"
14075
+ "An array of results for the Content Moderation model, if it is enabled.\nSee [Content moderation](https://www.assemblyai.com/docs/content-moderation) for more information.\n"
13314
14076
  ),
13315
14077
  custom_spelling: import_zod3.z.array(
13316
14078
  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"),
14079
+ from: import_zod3.z.array(import_zod3.z.string()).describe("Words or phrases to replace"),
13318
14080
  to: import_zod3.z.string().describe("Word to replace with")
13319
14081
  }).describe(
13320
14082
  "Object containing words or phrases to replace, and the word or phrase to replace with"
13321
14083
  )
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
14084
  ).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"
14085
+ "Customize how words are spelled and formatted using to and from values. See [Custom Spelling](https://www.assemblyai.com/docs/pre-recorded-audio/custom-spelling) for more details."
13376
14086
  ),
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"
14087
+ disfluencies: import_zod3.z.boolean().nullish().describe(
14088
+ 'Transcribe [Filler Words](https://www.assemblyai.com/docs/pre-recorded-audio/filler-words), like "umm", in your media file; can be true or false'
14089
+ ),
14090
+ domain: import_zod3.z.string().nullish().describe(
14091
+ 'The domain-specific model applied to the transcript. When set to `"medical-v1"`, [Medical Mode](https://www.assemblyai.com/docs/pre-recorded-audio/medical-mode) was used to improve accuracy for medical terminology.\n'
13379
14092
  ),
13380
14093
  entities: import_zod3.z.array(
13381
14094
  import_zod3.z.object({
@@ -13434,61 +14147,49 @@ var getTranscriptResponse = import_zod3.z.object({
13434
14147
  )
13435
14148
  }).describe("A detected entity")
13436
14149
  ).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"
14150
+ "An array of results for the Entity Detection model, if it is enabled.\nSee [Entity detection](https://www.assemblyai.com/docs/speech-understanding/entity-detection) for more information.\n"
13441
14151
  ),
13442
- throttled: import_zod3.z.boolean().nullish().describe(
13443
- "True while a request is throttled and false when a request is no longer throttled"
14152
+ entity_detection: import_zod3.z.boolean().nullish().describe(
14153
+ "Whether [Entity Detection](https://www.assemblyai.com/docs/speech-understanding/entity-detection) is enabled, can be true or false"
13444
14154
  ),
13445
14155
  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({
14156
+ filter_profanity: import_zod3.z.boolean().nullish().describe(
14157
+ "Whether [Profanity Filtering](https://www.assemblyai.com/docs/profanity-filtering) is enabled, either true or false"
14158
+ ),
14159
+ format_text: import_zod3.z.boolean().nullish().describe(
14160
+ "Whether [Text Formatting](https://www.assemblyai.com/docs/pre-recorded-audio) is enabled, either true or false"
14161
+ ),
14162
+ iab_categories: import_zod3.z.boolean().nullish().describe(
14163
+ "Whether [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection) is enabled, can be true or false"
14164
+ ),
14165
+ iab_categories_result: import_zod3.z.object({
14166
+ status: import_zod3.z.enum(["success", "unavailable"]).describe("Either success, or unavailable in the rare case that the model failed"),
14167
+ results: import_zod3.z.array(
14168
+ import_zod3.z.object({
14169
+ text: import_zod3.z.string().describe("The text in the transcript in which a detected topic occurs"),
14170
+ labels: import_zod3.z.array(
14171
+ import_zod3.z.object({
14172
+ relevance: import_zod3.z.number().describe("How relevant the detected topic is of a detected topic"),
14173
+ label: import_zod3.z.string().describe(
14174
+ "The IAB taxonomical label for the label of the detected topic, where > denotes supertopic/subtopic relationship"
14175
+ )
14176
+ })
14177
+ ).optional().describe("An array of detected topics in the text"),
14178
+ timestamp: import_zod3.z.object({
14179
+ start: import_zod3.z.number().describe("The start time in milliseconds"),
14180
+ end: import_zod3.z.number().describe("The end time in milliseconds")
14181
+ }).optional().describe("Timestamp containing a start and end property in milliseconds")
14182
+ }).describe("The result of the topic detection model")
14183
+ ).describe("An array of results for the Topic Detection model"),
14184
+ summary: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.number()).describe("The overall relevance of topic to the entire audio file")
14185
+ }).describe(
14186
+ "The result of the Topic Detection model, if it is enabled.\nSee [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection) for more information.\n"
14187
+ ).or(import_zod3.z.null()).optional().describe(
14188
+ "The result of the Topic Detection model, if it is enabled.\nSee [Topic Detection](https://www.assemblyai.com/docs/speech-understanding/topic-detection) for more information.\n"
14189
+ ),
13488
14190
  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."
14191
+ keyterms_prompt: import_zod3.z.array(import_zod3.z.string()).optional().describe(
14192
+ "Improve accuracy with up to 200 (for Universal-2) or 1000 (for Universal-3 Pro) domain-specific words or phrases (maximum 6 words per phrase). See [Keyterms Prompting](https://www.assemblyai.com/docs/pre-recorded-audio/keyterms-prompting) for more details.\n"
13492
14193
  ),
13493
14194
  language_code: import_zod3.z.enum([
13494
14195
  "en",
@@ -13593,138 +14294,175 @@ var deleteTranscriptResponse = import_zod3.z.object({
13593
14294
  "cy",
13594
14295
  "yi",
13595
14296
  "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"
14297
+ ]).optional().describe(
14298
+ "The language of your audio file. Possible values are found in [Supported Languages](https://www.assemblyai.com/docs/pre-recorded-audio/supported-languages).\nThe default value is 'en_us'.\n"
13629
14299
  ),
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
- })
14300
+ language_codes: import_zod3.z.array(
14301
+ import_zod3.z.enum([
14302
+ "en",
14303
+ "en_au",
14304
+ "en_uk",
14305
+ "en_us",
14306
+ "es",
14307
+ "fr",
14308
+ "de",
14309
+ "it",
14310
+ "pt",
14311
+ "nl",
14312
+ "af",
14313
+ "sq",
14314
+ "am",
14315
+ "ar",
14316
+ "hy",
14317
+ "as",
14318
+ "az",
14319
+ "ba",
14320
+ "eu",
14321
+ "be",
14322
+ "bn",
14323
+ "bs",
14324
+ "br",
14325
+ "bg",
14326
+ "my",
14327
+ "ca",
14328
+ "zh",
14329
+ "hr",
14330
+ "cs",
14331
+ "da",
14332
+ "et",
14333
+ "fo",
14334
+ "fi",
14335
+ "gl",
14336
+ "ka",
14337
+ "el",
14338
+ "gu",
14339
+ "ht",
14340
+ "ha",
14341
+ "haw",
14342
+ "he",
14343
+ "hi",
14344
+ "hu",
14345
+ "is",
14346
+ "id",
14347
+ "ja",
14348
+ "jw",
14349
+ "kn",
14350
+ "kk",
14351
+ "km",
14352
+ "ko",
14353
+ "lo",
14354
+ "la",
14355
+ "lv",
14356
+ "ln",
14357
+ "lt",
14358
+ "lb",
14359
+ "mk",
14360
+ "mg",
14361
+ "ms",
14362
+ "ml",
14363
+ "mt",
14364
+ "mi",
14365
+ "mr",
14366
+ "mn",
14367
+ "ne",
14368
+ "no",
14369
+ "nn",
14370
+ "oc",
14371
+ "pa",
14372
+ "ps",
14373
+ "fa",
14374
+ "pl",
14375
+ "ro",
14376
+ "ru",
14377
+ "sa",
14378
+ "sr",
14379
+ "sn",
14380
+ "sd",
14381
+ "si",
14382
+ "sk",
14383
+ "sl",
14384
+ "so",
14385
+ "su",
14386
+ "sw",
14387
+ "sv",
14388
+ "tl",
14389
+ "tg",
14390
+ "ta",
14391
+ "tt",
14392
+ "te",
14393
+ "th",
14394
+ "bo",
14395
+ "tr",
14396
+ "tk",
14397
+ "uk",
14398
+ "ur",
14399
+ "uz",
14400
+ "vi",
14401
+ "cy",
14402
+ "yi",
14403
+ "yo"
14404
+ ]).describe(
14405
+ "The language of your audio file. Possible values are found in [Supported Languages](https://www.assemblyai.com/docs/pre-recorded-audio/supported-languages).\nThe default value is 'en_us'.\n"
14406
+ )
13657
14407
  ).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."
14408
+ "The language codes of your audio file. Used for [Code switching](/docs/speech-to-text/pre-recorded-audio/code-switching)\nOne of the values specified must be `en`.\n"
13672
14409
  ),
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"
14410
+ language_confidence: import_zod3.z.number().nullable().describe(
14411
+ "The confidence score for the detected language, between 0.0 (low confidence) and 1.0 (high confidence). See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details."
13675
14412
  ),
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"
14413
+ language_confidence_threshold: import_zod3.z.number().nullable().describe(
14414
+ "The confidence threshold for the automatically detected language.\nAn error will be returned if the language confidence is below this threshold.\nSee [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.\n"
13678
14415
  ),
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"
14416
+ language_detection: import_zod3.z.boolean().nullish().describe(
14417
+ "Whether [Automatic language detection](/docs/pre-recorded-audio/automatic-language-detection) is enabled, either true or false"
13682
14418
  ),
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"
14419
+ language_detection_options: import_zod3.z.object({
14420
+ expected_languages: import_zod3.z.array(import_zod3.z.string()).optional().describe(
14421
+ 'List of languages expected in the audio file. Defaults to `["all"]` when unspecified. See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.'
14422
+ ),
14423
+ fallback_language: import_zod3.z.string().default(deleteTranscriptResponseLanguageDetectionOptionsFallbackLanguageDefault).describe(
14424
+ 'If the detected language of the audio file is not in the list of expected languages, the `fallback_language` is used. Specify `["auto"]` to let our model choose the fallback language from `expected_languages` with the highest confidence score. See [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection) for more details.\n'
14425
+ ),
14426
+ code_switching: import_zod3.z.boolean().optional().describe(
14427
+ "Whether [code switching](/docs/speech-to-text/pre-recorded-audio/code-switching) should be detected.\n"
14428
+ ),
14429
+ code_switching_confidence_threshold: import_zod3.z.number().default(
14430
+ deleteTranscriptResponseLanguageDetectionOptionsCodeSwitchingConfidenceThresholdDefault
14431
+ ).describe(
14432
+ "The confidence threshold for [code switching](/docs/speech-to-text/pre-recorded-audio/code-switching) detection. If the code switching confidence is below this threshold, the transcript will be processed in the language with the highest `language_detection_confidence` score.\n"
14433
+ )
14434
+ }).optional().describe(
14435
+ "Specify options for [Automatic Language Detection](https://www.assemblyai.com/docs/pre-recorded-audio/language-detection)."
13706
14436
  ),
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"
14437
+ multichannel: import_zod3.z.boolean().nullish().describe(
14438
+ "Whether [Multichannel transcription](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) was enabled in the transcription request, either true or false"
13709
14439
  ),
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"
14440
+ prompt: import_zod3.z.string().optional().describe(
14441
+ "Provide natural language prompting of up to 1,500 words of contextual information to the model. See the [Prompting Guide](https://www.assemblyai.com/docs/pre-recorded-audio/prompting) for best practices.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n"
13712
14442
  ),
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"
14443
+ punctuate: import_zod3.z.boolean().nullish().describe(
14444
+ "Whether [Automatic Punctuation](https://www.assemblyai.com/docs/pre-recorded-audio) is enabled, either true or false"
13717
14445
  ),
13718
14446
  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"
14447
+ "Whether [PII Redaction](https://www.assemblyai.com/docs/pii-redaction) is enabled, either true or false"
13720
14448
  ),
13721
14449
  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"
14450
+ "Whether a redacted version of the audio file was generated,\neither true or false. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more information.\n"
14451
+ ),
14452
+ redact_pii_audio_options: import_zod3.z.object({
14453
+ return_redacted_no_speech_audio: import_zod3.z.boolean().optional().describe(
14454
+ "By default, audio redaction provides redacted audio URLs only when speech is detected. However, if your use-case specifically requires redacted audio files even for silent audio files without any dialogue, you can opt to receive these URLs by setting this parameter to `true`."
14455
+ ),
14456
+ override_audio_redaction_method: import_zod3.z.enum(["silence"]).optional().describe(
14457
+ "Specify the method used to redact audio. By default, redacted audio uses a beep sound. Set to `silence` to replace PII with silence instead of a beep."
14458
+ )
14459
+ }).optional().describe(
14460
+ "The options for PII-redacted audio, if redact_pii_audio is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more information.\n"
13723
14461
  ),
13724
14462
  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."
14463
+ "Controls the filetype of the audio created by redact_pii_audio. Currently supports mp3 (default) and wav. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more details."
13726
14464
  ).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"
14465
+ "The audio quality of the PII-redacted audio file, if redact_pii_audio is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/pii-redaction#request-for-redacted-audio) for more information.\n"
13728
14466
  ),
13729
14467
  redact_pii_policies: import_zod3.z.array(
13730
14468
  import_zod3.z.enum([
@@ -13774,271 +14512,256 @@ var deleteTranscriptResponse = import_zod3.z.object({
13774
14512
  "zodiac_sign"
13775
14513
  ]).describe("The type of PII to redact")
13776
14514
  ).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"
14515
+ "The list of PII Redaction policies that were enabled, if PII Redaction is enabled.\nSee [PII redaction](https://www.assemblyai.com/docs/pii-redaction) for more information.\n"
13778
14516
  ),
13779
14517
  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.'
14518
+ "The replacement logic for detected PII, can be `entity_name` or `hash`. See [PII redaction](https://www.assemblyai.com/docs/pii-redaction) for more details."
14519
+ ),
14520
+ sentiment_analysis: import_zod3.z.boolean().nullish().describe(
14521
+ "Whether [Sentiment Analysis](https://www.assemblyai.com/docs/speech-understanding/sentiment-analysis) is enabled, can be true or false"
14522
+ ),
14523
+ sentiment_analysis_results: import_zod3.z.array(
14524
+ import_zod3.z.object({
14525
+ text: import_zod3.z.string().describe("The transcript of the sentence"),
14526
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, of the sentence"),
14527
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, of the sentence"),
14528
+ sentiment: import_zod3.z.enum(["POSITIVE", "NEUTRAL", "NEGATIVE"]),
14529
+ confidence: import_zod3.z.number().describe(
14530
+ "The confidence score for the detected sentiment of the sentence, from 0 to 1"
14531
+ ),
14532
+ channel: import_zod3.z.string().nullish().describe(
14533
+ "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
14534
+ ),
14535
+ speaker: import_zod3.z.string().nullable().describe(
14536
+ "The speaker of the sentence if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
14537
+ )
14538
+ }).describe("The result of the Sentiment Analysis model")
14539
+ ).nullish().describe(
14540
+ "An array of results for the Sentiment Analysis model, if it is enabled.\nSee [Sentiment Analysis](https://www.assemblyai.com/docs/speech-understanding/sentiment-analysis) for more information.\n"
13781
14541
  ),
13782
14542
  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"
14543
+ "Whether [Speaker diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, can be true or false"
13784
14544
  ),
13785
14545
  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."
14546
+ "Tell the speaker label model how many speakers it should attempt to identify. See [Set number of speakers expected](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization#set-number-of-speakers-expected) for more details."
13787
14547
  ),
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"
14548
+ speech_model_used: import_zod3.z.string().optional().describe(
14549
+ "The speech model to use for the transcription. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models."
13790
14550
  ),
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"
14551
+ speech_models: import_zod3.z.array(
14552
+ import_zod3.z.string().describe(
14553
+ "The speech model to use for the transcription. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models."
14554
+ )
14555
+ ).nullish().describe(
14556
+ "List multiple speech models in priority order, allowing our system to automatically route your audio to the best available option. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models and routing behavior.\n"
14557
+ ),
14558
+ speech_threshold: import_zod3.z.number().nullish().describe(
14559
+ "Defaults to null. Reject audio files that contain less than this fraction of speech.\nValid values are in the range [0, 1] inclusive. See [Speech Threshold](https://www.assemblyai.com/docs/speech-threshold) for more details.\n"
14560
+ ),
14561
+ speech_understanding: import_zod3.z.object({
14562
+ request: import_zod3.z.object({
14563
+ translation: import_zod3.z.object({
14564
+ target_languages: import_zod3.z.array(import_zod3.z.string()).describe(
14565
+ 'List of target language codes (e.g., `["es", "de"]`). See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for supported languages.'
13812
14566
  ),
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")
14567
+ formal: import_zod3.z.boolean().default(deleteTranscriptResponseSpeechUnderstandingRequestTranslationFormalDefault).describe(
14568
+ "Use formal language style. See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for more details."
14569
+ ),
14570
+ match_original_utterance: import_zod3.z.boolean().optional().describe(
14571
+ "When enabled with Speaker Labels, returns translated text in the utterances array. Each utterance will include a `translated_texts` key containing translations for each target language."
14572
+ )
13819
14573
  })
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'
14574
+ }).describe(
14575
+ "Request body for [Translation](https://www.assemblyai.com/docs/speech-understanding/translation)."
14576
+ ).or(
14577
+ import_zod3.z.object({
14578
+ speaker_identification: import_zod3.z.object({
14579
+ speaker_type: import_zod3.z.enum(["role", "name"]).describe(
14580
+ "Type of speaker identification. See [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification) for details on each type."
14581
+ ),
14582
+ known_values: import_zod3.z.array(import_zod3.z.string()).optional().describe(
14583
+ 'Required if speaker_type is "role". Each value must be 35 characters or less.'
14584
+ ),
14585
+ speakers: import_zod3.z.array(
14586
+ import_zod3.z.object({
14587
+ role: import_zod3.z.string().optional().describe(
14588
+ 'The role of the speaker. Required when `speaker_type` is "role".'
14589
+ ),
14590
+ name: import_zod3.z.string().optional().describe(
14591
+ 'The name of the speaker. Required when `speaker_type` is "name".'
14592
+ ),
14593
+ description: import_zod3.z.string().optional().describe(
14594
+ "A description of the speaker to help the model identify them based on conversational context."
14595
+ )
14596
+ })
14597
+ ).optional().describe(
14598
+ "An array of speaker objects with metadata to improve identification accuracy. Each object should include a `role` or `name` (depending on `speaker_type`) and an optional `description` to help the model identify the speaker. You can also include any additional custom properties (e.g., `company`, `title`) to provide more context. Use this as an alternative to `known_values` when you want to provide additional context about each speaker."
14599
+ )
14600
+ })
14601
+ }).describe(
14602
+ "Request body for [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification)."
13825
14603
  )
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(),
14604
+ ).or(
14605
+ import_zod3.z.object({
14606
+ custom_formatting: import_zod3.z.object({
14607
+ date: import_zod3.z.string().optional().describe(
14608
+ 'Date format pattern (e.g., `"mm/dd/yyyy"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
14609
+ ),
14610
+ phone_number: import_zod3.z.string().optional().describe(
14611
+ 'Phone number format pattern (e.g., `"(xxx)xxx-xxxx"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
14612
+ ),
14613
+ email: import_zod3.z.string().optional().describe(
14614
+ 'Email format pattern (e.g., `"username@domain.com"`). See [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting) for more details.'
14615
+ )
14616
+ })
14617
+ }).describe(
14618
+ "Request body for [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting)."
14619
+ )
14620
+ ).optional(),
14621
+ response: import_zod3.z.object({
14622
+ translation: import_zod3.z.object({
14623
+ status: import_zod3.z.string().optional()
14624
+ }).optional()
14625
+ }).or(
13831
14626
  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)
14627
+ speaker_identification: import_zod3.z.object({
14628
+ mapping: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional().describe(
14629
+ 'A mapping of the original generic speaker labels (e.g., "A", "B") to the identified speaker names or roles.'
14630
+ ),
14631
+ status: import_zod3.z.string().optional()
14632
+ }).optional()
13835
14633
  })
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(
14634
+ ).or(
13850
14635
  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"
14636
+ custom_formatting: import_zod3.z.object({
14637
+ mapping: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional(),
14638
+ formatted_text: import_zod3.z.string().optional()
14639
+ }).optional()
14640
+ })
14641
+ ).optional()
14642
+ }).optional().describe(
14643
+ "Speech understanding tasks like [Translation](https://www.assemblyai.com/docs/speech-understanding/translation), [Speaker Identification](https://www.assemblyai.com/docs/speech-understanding/speaker-identification), and [Custom Formatting](https://www.assemblyai.com/docs/speech-understanding/custom-formatting). See the task-specific docs for available options and configuration.\n"
13889
14644
  ),
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"
14645
+ status: import_zod3.z.enum(["queued", "processing", "completed", "error"]).describe(
14646
+ "The status of your transcript. Possible values are queued, processing, completed, or error."
13893
14647
  ),
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
14648
  summarization: import_zod3.z.boolean().describe(
13906
- "Whether [Summarization](https://www.assemblyai.com/docs/models/summarization) is enabled, either true or false"
14649
+ "Whether [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization) is enabled, either true or false. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details.\n\nNote: This parameter is only supported for the Universal-2 model.\n"
13907
14650
  ),
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"
14651
+ summary: import_zod3.z.string().nullish().describe(
14652
+ "The generated summary of the media file, if [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization) is enabled. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details."
13910
14653
  ),
13911
14654
  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"
14655
+ "The Summarization model used to generate the summary,\nif [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization#summary-models) is enabled. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details.\n"
13913
14656
  ),
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"
14657
+ summary_type: import_zod3.z.string().nullish().describe(
14658
+ "The type of summary generated, if [Summarization](https://www.assemblyai.com/docs/speech-understanding/summarization#summary-types) is enabled. Deprecated - use [LLM Gateway](https://www.assemblyai.com/docs/llm-gateway/overview) instead for more flexible summaries. See the [updated Summarization page](https://www.assemblyai.com/docs/speech-understanding/summarization) for details."
13916
14659
  ),
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"
14660
+ remove_audio_tags: import_zod3.z.enum(["all"]).describe(
14661
+ "Whether [audio event tags](https://www.assemblyai.com/docs/pre-recorded-audio/universal-3-pro#audio-event-tags) were removed from the transcript text.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n"
14662
+ ).or(import_zod3.z.null()).optional().describe(
14663
+ "Whether [audio event tags](https://www.assemblyai.com/docs/pre-recorded-audio/universal-3-pro#audio-event-tags) were removed from the transcript text.\n\nNote: This parameter is only supported for the Universal-3 Pro model.\n"
13921
14664
  ),
13922
- sentiment_analysis_results: import_zod3.z.array(
14665
+ temperature: import_zod3.z.number().nullish().describe(
14666
+ "The temperature that was used for the model's response. See the [Prompting Guide](https://www.assemblyai.com/docs/pre-recorded-audio/prompting) for more details.\n\nNote: This parameter can only be used with the Universal-3 Pro model.\n"
14667
+ ),
14668
+ text: import_zod3.z.string().nullish().describe("The textual transcript of your media file"),
14669
+ throttled: import_zod3.z.boolean().nullish().describe(
14670
+ "True while a request is throttled and false when a request is no longer throttled"
14671
+ ),
14672
+ utterances: import_zod3.z.array(
13923
14673
  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
- ),
14674
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this utterance"),
14675
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, of the utterance in the audio file"),
14676
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, of the utterance in the audio file"),
14677
+ text: import_zod3.z.string().describe("The text for this utterance"),
14678
+ words: import_zod3.z.array(
14679
+ import_zod3.z.object({
14680
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
14681
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
14682
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
14683
+ text: import_zod3.z.string().describe("The text of the word"),
14684
+ channel: import_zod3.z.string().nullish().describe(
14685
+ "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
14686
+ ),
14687
+ speaker: import_zod3.z.string().nullable().describe(
14688
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
14689
+ )
14690
+ })
14691
+ ).describe("The words in the utterance."),
13931
14692
  channel: import_zod3.z.string().nullish().describe(
13932
14693
  "The channel of this utterance. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13933
14694
  ),
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"
14695
+ speaker: import_zod3.z.string().describe(
14696
+ 'The speaker of this utterance, where each speaker is assigned a sequential capital letter - e.g. "A" for Speaker A, "B" for Speaker B, etc.'
14697
+ ),
14698
+ translated_texts: import_zod3.z.record(import_zod3.z.string(), import_zod3.z.string()).optional().describe(
14699
+ 'Translations keyed by language code (e.g., `{"es": "Texto traducido", "de": "\xDCbersetzter Text"}`). Only present when `match_original_utterance` is enabled with translation.'
13936
14700
  )
13937
- }).describe("The result of the Sentiment Analysis model")
14701
+ })
13938
14702
  ).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"
14703
+ "When multichannel or speaker_labels is enabled, a list of turn-by-turn utterance objects.\nSee [Speaker diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) and [Multichannel transcription](https://www.assemblyai.com/docs/pre-recorded-audio/multichannel) for more information.\n"
13940
14704
  ),
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"
14705
+ webhook_auth: import_zod3.z.boolean().describe(
14706
+ "Whether [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) authentication details were provided"
13943
14707
  ),
13944
- entities: import_zod3.z.array(
14708
+ webhook_auth_header_name: import_zod3.z.string().nullish().describe(
14709
+ "The header name to be sent with the transcript completed or failed [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) requests"
14710
+ ),
14711
+ webhook_status_code: import_zod3.z.number().nullish().describe(
14712
+ "The status code we received from your server when delivering the transcript completed or failed [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) request, if a webhook URL was provided"
14713
+ ),
14714
+ webhook_url: import_zod3.z.string().nullish().describe(
14715
+ "The URL to which we send [webhook](https://www.assemblyai.com/docs/deployment/webhooks-for-pre-recorded-audio) requests.\n"
14716
+ ),
14717
+ words: import_zod3.z.array(
13945
14718
  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"
14719
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
14720
+ start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
14721
+ end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
14722
+ text: import_zod3.z.string().describe("The text of the word"),
14723
+ channel: import_zod3.z.string().nullish().describe(
14724
+ "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
13995
14725
  ),
13996
- end: import_zod3.z.number().describe(
13997
- "The ending time, in milliseconds, for the detected entity in the audio file"
14726
+ speaker: import_zod3.z.string().nullable().describe(
14727
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
13998
14728
  )
13999
- }).describe("A detected entity")
14729
+ })
14000
14730
  ).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")
14731
+ "An array of temporally-sequential word objects, one for each word in the transcript.\n"
14732
+ ),
14733
+ acoustic_model: import_zod3.z.string().describe("This parameter does not currently have any functionality attached to it."),
14734
+ custom_topics: import_zod3.z.boolean().nullish().describe("This parameter does not currently have any functionality attached to it."),
14735
+ language_model: import_zod3.z.string().describe("This parameter does not currently have any functionality attached to it."),
14736
+ speech_model: import_zod3.z.string().describe(
14737
+ "The speech model to use for the transcription. See [Model Selection](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model) for available models."
14738
+ ).or(import_zod3.z.null()).describe(
14739
+ "This parameter has been replaced with the `speech_models` parameter, learn more about the `speech_models` parameter [here](https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model).\n"
14740
+ ),
14741
+ speed_boost: import_zod3.z.boolean().nullish().describe("This parameter does not currently have any functionality attached to it."),
14742
+ topics: import_zod3.z.array(import_zod3.z.string()).optional().describe("This parameter does not currently have any functionality attached to it."),
14743
+ translated_texts: import_zod3.z.object({
14744
+ language_code: import_zod3.z.string().optional().describe("Translated text for this language code")
14745
+ }).optional().describe(
14746
+ "Translated text keyed by language code. See [Translation](https://www.assemblyai.com/docs/speech-understanding/translation) for more details."
14747
+ )
14012
14748
  }).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
14749
  var getTranscriptSentencesParams = import_zod3.z.object({
14021
14750
  transcript_id: import_zod3.z.string().describe("ID of the transcript")
14022
14751
  });
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
14752
  var getTranscriptSentencesResponse = import_zod3.z.object({
14030
14753
  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"),
14754
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript"),
14032
14755
  audio_duration: import_zod3.z.number().describe("The duration of the audio file in seconds"),
14033
14756
  sentences: import_zod3.z.array(
14034
14757
  import_zod3.z.object({
14035
14758
  text: import_zod3.z.string().describe("The transcript of the sentence"),
14036
14759
  start: import_zod3.z.number().describe("The starting time, in milliseconds, for the sentence"),
14037
14760
  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"),
14761
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this sentence"),
14039
14762
  words: import_zod3.z.array(
14040
14763
  import_zod3.z.object({
14041
- confidence: import_zod3.z.number().min(getTranscriptSentencesResponseSentencesItemWordsItemConfidenceMin).max(getTranscriptSentencesResponseSentencesItemWordsItemConfidenceMax).describe("The confidence score for the transcript of this word"),
14764
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
14042
14765
  start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
14043
14766
  end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
14044
14767
  text: import_zod3.z.string().describe("The text of the word"),
@@ -14046,7 +14769,7 @@ var getTranscriptSentencesResponse = import_zod3.z.object({
14046
14769
  "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
14047
14770
  ),
14048
14771
  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"
14772
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
14050
14773
  )
14051
14774
  })
14052
14775
  ).describe("An array of words in the sentence"),
@@ -14054,7 +14777,7 @@ var getTranscriptSentencesResponse = import_zod3.z.object({
14054
14777
  "The channel of the sentence. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
14055
14778
  ),
14056
14779
  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"
14780
+ "The speaker of the sentence if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
14058
14781
  )
14059
14782
  })
14060
14783
  ).describe("An array of sentences in the transcript")
@@ -14062,25 +14785,19 @@ var getTranscriptSentencesResponse = import_zod3.z.object({
14062
14785
  var getTranscriptParagraphsParams = import_zod3.z.object({
14063
14786
  transcript_id: import_zod3.z.string().describe("ID of the transcript")
14064
14787
  });
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
14788
  var getTranscriptParagraphsResponse = import_zod3.z.object({
14072
14789
  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"),
14790
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript"),
14074
14791
  audio_duration: import_zod3.z.number().describe("The duration of the audio file in seconds"),
14075
14792
  paragraphs: import_zod3.z.array(
14076
14793
  import_zod3.z.object({
14077
14794
  text: import_zod3.z.string().describe("The transcript of the paragraph"),
14078
14795
  start: import_zod3.z.number().describe("The starting time, in milliseconds, of the paragraph"),
14079
14796
  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"),
14797
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this paragraph"),
14081
14798
  words: import_zod3.z.array(
14082
14799
  import_zod3.z.object({
14083
- confidence: import_zod3.z.number().min(getTranscriptParagraphsResponseParagraphsItemWordsItemConfidenceMin).max(getTranscriptParagraphsResponseParagraphsItemWordsItemConfidenceMax).describe("The confidence score for the transcript of this word"),
14800
+ confidence: import_zod3.z.number().describe("The confidence score for the transcript of this word"),
14084
14801
  start: import_zod3.z.number().describe("The starting time, in milliseconds, for the word"),
14085
14802
  end: import_zod3.z.number().describe("The ending time, in milliseconds, for the word"),
14086
14803
  text: import_zod3.z.string().describe("The text of the word"),
@@ -14088,13 +14805,28 @@ var getTranscriptParagraphsResponse = import_zod3.z.object({
14088
14805
  "The channel of the word. The left and right channels are channels 1 and 2. Additional channels increment the channel number sequentially."
14089
14806
  ),
14090
14807
  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"
14808
+ "The speaker of the word if [Speaker Diarization](https://www.assemblyai.com/docs/pre-recorded-audio/speaker-diarization) is enabled, else null"
14092
14809
  )
14093
14810
  })
14094
14811
  ).describe("An array of words in the paragraph")
14095
14812
  })
14096
14813
  ).describe("An array of paragraphs in the transcript")
14097
14814
  });
14815
+ var getSubtitlesParams = import_zod3.z.object({
14816
+ transcript_id: import_zod3.z.string().describe("ID of the transcript"),
14817
+ subtitle_format: import_zod3.z.enum(["srt", "vtt"]).describe("The format of the captions")
14818
+ });
14819
+ var getSubtitlesQueryParams = import_zod3.z.object({
14820
+ chars_per_caption: import_zod3.z.number().optional().describe("The maximum number of characters per caption")
14821
+ });
14822
+ var getSubtitlesResponse = import_zod3.z.object({});
14823
+ var getRedactedAudioParams = import_zod3.z.object({
14824
+ transcript_id: import_zod3.z.string().describe("ID of the transcript")
14825
+ });
14826
+ var getRedactedAudioResponse = import_zod3.z.object({
14827
+ status: import_zod3.z.enum(["redacted_audio_ready"]).describe("The status of the redacted audio"),
14828
+ redacted_audio_url: import_zod3.z.string().describe("The URL of the redacted audio file")
14829
+ });
14098
14830
  var wordSearchParams = import_zod3.z.object({
14099
14831
  transcript_id: import_zod3.z.string().describe("ID of the transcript")
14100
14832
  });
@@ -14111,7 +14843,7 @@ var wordSearchResponse = import_zod3.z.object({
14111
14843
  text: import_zod3.z.string().describe("The matched word"),
14112
14844
  count: import_zod3.z.number().describe("The total amount of times the word is in the transcript"),
14113
14845
  timestamps: import_zod3.z.array(
14114
- import_zod3.z.array(import_zod3.z.number().describe("Timestamp in milliseconds")).describe(
14846
+ import_zod3.z.array(import_zod3.z.number()).describe(
14115
14847
  "An array of timestamps structured as [`start_time`, `end_time`] in milliseconds"
14116
14848
  )
14117
14849
  ).describe("An array of timestamps"),
@@ -14121,205 +14853,6 @@ var wordSearchResponse = import_zod3.z.object({
14121
14853
  })
14122
14854
  ).describe("The matches of the search")
14123
14855
  });
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
14856
 
14324
14857
  // src/generated/assemblyai/streaming-types.zod.ts
14325
14858
  var import_zod4 = require("zod");
@@ -14334,25 +14867,37 @@ var streamingTranscriberParams = import_zod4.z.object({
14334
14867
  enableExtraSessionInformation: import_zod4.z.boolean().optional().describe(
14335
14868
  "Set to true to receive the SessionInformation message before the session ends. Defaults to false."
14336
14869
  ),
14870
+ domain: import_zod4.z.string().optional().describe(
14871
+ 'Enable domain-specific transcription models to improve accuracy for specialized terminology. Set to `"medical-v1"` to enable [Medical Mode](https://www.assemblyai.com/docs/streaming/medical-mode) for improved accuracy of medical terms such as medications, procedures, conditions, and dosages. Supported languages: English (`en`), Spanish (`es`), German (`de`), French (`fr`). If used with an unsupported language, the parameter is ignored and a warning is returned.'
14872
+ ),
14337
14873
  endOfTurnConfidenceThreshold: import_zod4.z.number().optional().describe("From SDK v3"),
14338
14874
  minEndOfTurnSilenceWhenConfident: import_zod4.z.number().optional().describe("From SDK v3"),
14875
+ minTurnSilence: import_zod4.z.number().optional().describe("From SDK v3"),
14339
14876
  maxTurnSilence: import_zod4.z.number().optional().describe("From SDK v3"),
14340
14877
  vadThreshold: import_zod4.z.number().optional().describe("From SDK v3"),
14341
14878
  formatTurns: import_zod4.z.boolean().optional().describe("From SDK v3"),
14342
14879
  filterProfanity: import_zod4.z.boolean().optional().describe("From SDK v3"),
14343
14880
  keyterms: import_zod4.z.array(import_zod4.z.string()).optional().describe("From SDK v3"),
14344
14881
  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"),
14882
+ prompt: import_zod4.z.string().optional().describe("From SDK v3"),
14883
+ speechModel: import_zod4.z.enum(["universal-streaming-english", "universal-streaming-multilingual"]).describe("From SDK v3"),
14346
14884
  languageDetection: import_zod4.z.boolean().optional().describe("From SDK v3"),
14347
- inactivityTimeout: import_zod4.z.number().optional().describe("From SDK v3")
14885
+ inactivityTimeout: import_zod4.z.number().optional().describe("From SDK v3"),
14886
+ speakerLabels: import_zod4.z.boolean().optional().describe("From SDK v3"),
14887
+ maxSpeakers: import_zod4.z.number().optional().describe("From SDK v3"),
14888
+ llmGateway: import_zod4.z.unknown().optional().describe("From SDK v3")
14348
14889
  });
14349
14890
  var streamingUpdateConfigParams = import_zod4.z.object({
14350
14891
  end_utterance_silence_threshold: import_zod4.z.number().min(0).max(2e4).optional().describe("The duration threshold in milliseconds"),
14351
14892
  end_of_turn_confidence_threshold: import_zod4.z.number().optional().describe("From SDK v3"),
14352
14893
  min_end_of_turn_silence_when_confident: import_zod4.z.number().optional().describe("From SDK v3"),
14894
+ min_turn_silence: import_zod4.z.number().optional().describe("From SDK v3"),
14353
14895
  max_turn_silence: import_zod4.z.number().optional().describe("From SDK v3"),
14354
14896
  vad_threshold: import_zod4.z.number().optional().describe("From SDK v3"),
14355
- format_turns: import_zod4.z.boolean().optional().describe("From SDK v3")
14897
+ format_turns: import_zod4.z.boolean().optional().describe("From SDK v3"),
14898
+ keyterms_prompt: import_zod4.z.array(import_zod4.z.string()).optional().describe("From SDK v3"),
14899
+ prompt: import_zod4.z.string().optional().describe("From SDK v3"),
14900
+ filter_profanity: import_zod4.z.boolean().optional().describe("From SDK v3")
14356
14901
  });
14357
14902
 
14358
14903
  // src/generated/gladia/api/gladiaControlAPI.zod.ts
@@ -40376,7 +40921,6 @@ var deleteTranscriptByIdResponse = import_zod13.z.any();
40376
40921
  createOpenAIWhisperAdapter,
40377
40922
  createSonioxAdapter,
40378
40923
  createSpeechmaticsAdapter,
40379
- createTemporaryToken,
40380
40924
  createTranscript,
40381
40925
  createTranscription,
40382
40926
  createVoiceRouter,