assemblyai 4.2.2 → 4.2.3

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.
@@ -991,6 +991,11 @@ export type SeverityScoreSummary = {
991
991
  /** Format: double */
992
992
  medium: number;
993
993
  };
994
+ /**
995
+ * @description The speech model to use for the transcription.
996
+ * @enum {string}
997
+ */
998
+ export type SpeechModel = "nano";
994
999
  /**
995
1000
  * @description 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.
996
1001
  * @enum {string}
@@ -1187,6 +1192,7 @@ export type TopicDetectionResult = {
1187
1192
  * @description A transcript object
1188
1193
  * @example {
1189
1194
  * "id": "9ea68fd3-f953-42c1-9742-976c447fb463",
1195
+ * "speech_model": null,
1190
1196
  * "language_model": "assemblyai_default",
1191
1197
  * "acoustic_model": "assemblyai_default",
1192
1198
  * "language_code": "en_us",
@@ -1989,6 +1995,11 @@ export type Transcript = {
1989
1995
  speaker_labels?: boolean | null;
1990
1996
  /** @description Tell the speaker label model how many speakers it should attempt to identify, up to 10. See [Speaker diarization](https://www.assemblyai.com/docs/models/speaker-diarization) for more details. */
1991
1997
  speakers_expected?: number | null;
1998
+ /**
1999
+ * @description The speech model used for the transcription. When `null`, the default model is used.
2000
+ * @default null
2001
+ */
2002
+ speech_model: SpeechModel | null;
1992
2003
  /**
1993
2004
  * Format: float
1994
2005
  * @description Defaults to null. Reject audio files that contain less than this fraction of speech.
@@ -2123,6 +2134,7 @@ export type TranscriptListItem = {
2123
2134
  /**
2124
2135
  * @description The parameters for creating a transcript
2125
2136
  * @example {
2137
+ * "speech_model": null,
2126
2138
  * "language_code": "en_us",
2127
2139
  * "punctuate": true,
2128
2140
  * "format_text": true,
@@ -2228,6 +2240,11 @@ export type TranscriptOptionalParams = {
2228
2240
  * @default null
2229
2241
  */
2230
2242
  speakers_expected?: number | null;
2243
+ /**
2244
+ * @description The speech model to use for the transcription. When `null`, the default model is used.
2245
+ * @default null
2246
+ */
2247
+ speech_model?: SpeechModel | null;
2231
2248
  /**
2232
2249
  * Format: float
2233
2250
  * @description Reject audio files that contain less than this fraction of speech.
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "assemblyai",
3
- "version": "4.2.2",
3
+ "version": "4.2.3",
4
4
  "description": "The AssemblyAI JavaScript SDK provides an easy-to-use interface for interacting with the AssemblyAI API, which supports async and real-time transcription, as well as the latest LeMUR models.",
5
5
  "engines": {
6
6
  "node": ">=18"
@@ -1102,6 +1102,12 @@ export type SeverityScoreSummary = {
1102
1102
  medium: number;
1103
1103
  };
1104
1104
 
1105
+ /**
1106
+ * @description The speech model to use for the transcription.
1107
+ * @enum {string}
1108
+ */
1109
+ export type SpeechModel = "nano";
1110
+
1105
1111
  /**
1106
1112
  * @description 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.
1107
1113
  * @enum {string}
@@ -1310,6 +1316,7 @@ export type TopicDetectionResult = {
1310
1316
  * @description A transcript object
1311
1317
  * @example {
1312
1318
  * "id": "9ea68fd3-f953-42c1-9742-976c447fb463",
1319
+ * "speech_model": null,
1313
1320
  * "language_model": "assemblyai_default",
1314
1321
  * "acoustic_model": "assemblyai_default",
1315
1322
  * "language_code": "en_us",
@@ -2112,6 +2119,11 @@ export type Transcript = {
2112
2119
  speaker_labels?: boolean | null;
2113
2120
  /** @description Tell the speaker label model how many speakers it should attempt to identify, up to 10. See [Speaker diarization](https://www.assemblyai.com/docs/models/speaker-diarization) for more details. */
2114
2121
  speakers_expected?: number | null;
2122
+ /**
2123
+ * @description The speech model used for the transcription. When `null`, the default model is used.
2124
+ * @default null
2125
+ */
2126
+ speech_model: SpeechModel | null;
2115
2127
  /**
2116
2128
  * Format: float
2117
2129
  * @description Defaults to null. Reject audio files that contain less than this fraction of speech.
@@ -2272,6 +2284,7 @@ export type TranscriptListItem = {
2272
2284
  /**
2273
2285
  * @description The parameters for creating a transcript
2274
2286
  * @example {
2287
+ * "speech_model": null,
2275
2288
  * "language_code": "en_us",
2276
2289
  * "punctuate": true,
2277
2290
  * "format_text": true,
@@ -2377,6 +2390,11 @@ export type TranscriptOptionalParams = {
2377
2390
  * @default null
2378
2391
  */
2379
2392
  speakers_expected?: number | null;
2393
+ /**
2394
+ * @description The speech model to use for the transcription. When `null`, the default model is used.
2395
+ * @default null
2396
+ */
2397
+ speech_model?: SpeechModel | null;
2380
2398
  /**
2381
2399
  * Format: float
2382
2400
  * @description Reject audio files that contain less than this fraction of speech.