sarvamai 1.1.3 → 1.1.4

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.
Files changed (43) hide show
  1. package/dist/cjs/BaseClient.js +2 -2
  2. package/dist/cjs/api/resources/speechToText/client/requests/SpeechToTextTranscriptionRequest.d.ts +14 -0
  3. package/dist/cjs/api/resources/speechToTextStreaming/types/SpeechToTextStreamingLanguageCode.d.ts +29 -1
  4. package/dist/cjs/api/resources/speechToTextStreaming/types/SpeechToTextStreamingLanguageCode.js +29 -1
  5. package/dist/cjs/api/resources/textToSpeechStreaming/types/TextToSpeechStreamingModel.d.ts +3 -3
  6. package/dist/cjs/api/resources/textToSpeechStreaming/types/TextToSpeechStreamingModel.js +3 -3
  7. package/dist/cjs/api/types/ConfigureConnection.d.ts +22 -34
  8. package/dist/cjs/api/types/ConfigureConnection.js +5 -19
  9. package/dist/cjs/api/types/SpeechToTextJobParameters.d.ts +15 -1
  10. package/dist/cjs/api/types/SpeechToTextLanguage.d.ts +19 -0
  11. package/dist/cjs/api/types/SpeechToTextLanguage.js +19 -0
  12. package/dist/cjs/api/types/SpeechToTextResponse.d.ts +13 -0
  13. package/dist/cjs/api/types/SpeechToTextTranscriptionData.d.ts +13 -0
  14. package/dist/cjs/api/types/SpeechToTextTranslateLanguage.d.ts +19 -0
  15. package/dist/cjs/api/types/SpeechToTextTranslateLanguage.js +19 -0
  16. package/dist/cjs/api/types/SpeechToTextTranslateResponse.d.ts +13 -0
  17. package/dist/cjs/api/types/SpeechToTextTranslateTranscriptionData.d.ts +13 -0
  18. package/dist/cjs/api/types/TextToSpeechModel.d.ts +1 -1
  19. package/dist/cjs/api/types/TextToSpeechModel.js +1 -1
  20. package/dist/cjs/version.d.ts +1 -1
  21. package/dist/cjs/version.js +1 -1
  22. package/dist/esm/BaseClient.mjs +2 -2
  23. package/dist/esm/api/resources/speechToText/client/requests/SpeechToTextTranscriptionRequest.d.mts +14 -0
  24. package/dist/esm/api/resources/speechToTextStreaming/types/SpeechToTextStreamingLanguageCode.d.mts +29 -1
  25. package/dist/esm/api/resources/speechToTextStreaming/types/SpeechToTextStreamingLanguageCode.mjs +29 -1
  26. package/dist/esm/api/resources/textToSpeechStreaming/types/TextToSpeechStreamingModel.d.mts +3 -3
  27. package/dist/esm/api/resources/textToSpeechStreaming/types/TextToSpeechStreamingModel.mjs +3 -3
  28. package/dist/esm/api/types/ConfigureConnection.d.mts +22 -34
  29. package/dist/esm/api/types/ConfigureConnection.mjs +5 -19
  30. package/dist/esm/api/types/SpeechToTextJobParameters.d.mts +15 -1
  31. package/dist/esm/api/types/SpeechToTextLanguage.d.mts +19 -0
  32. package/dist/esm/api/types/SpeechToTextLanguage.mjs +19 -0
  33. package/dist/esm/api/types/SpeechToTextResponse.d.mts +13 -0
  34. package/dist/esm/api/types/SpeechToTextTranscriptionData.d.mts +13 -0
  35. package/dist/esm/api/types/SpeechToTextTranslateLanguage.d.mts +19 -0
  36. package/dist/esm/api/types/SpeechToTextTranslateLanguage.mjs +19 -0
  37. package/dist/esm/api/types/SpeechToTextTranslateResponse.d.mts +13 -0
  38. package/dist/esm/api/types/SpeechToTextTranslateTranscriptionData.d.mts +13 -0
  39. package/dist/esm/api/types/TextToSpeechModel.d.mts +1 -1
  40. package/dist/esm/api/types/TextToSpeechModel.mjs +1 -1
  41. package/dist/esm/version.d.mts +1 -1
  42. package/dist/esm/version.mjs +1 -1
  43. package/package.json +1 -1
@@ -43,8 +43,8 @@ function normalizeClientOptions(options) {
43
43
  const headers = (0, headers_js_1.mergeHeaders)({
44
44
  "X-Fern-Language": "JavaScript",
45
45
  "X-Fern-SDK-Name": "sarvamai",
46
- "X-Fern-SDK-Version": "1.1.3",
47
- "User-Agent": "sarvamai/1.1.3",
46
+ "X-Fern-SDK-Version": "1.1.4",
47
+ "User-Agent": "sarvamai/1.1.4",
48
48
  "X-Fern-Runtime": core.RUNTIME.type,
49
49
  "X-Fern-Runtime-Version": core.RUNTIME.version,
50
50
  }, options === null || options === void 0 ? void 0 : options.headers);
@@ -59,6 +59,20 @@ export interface SpeechToTextTranscriptionRequest {
59
59
  * - `te-IN`: Telugu
60
60
  * - `en-IN`: English
61
61
  * - `gu-IN`: Gujarati
62
+ *
63
+ * **Additional Options (saaras:v3 only):**
64
+ * - `as-IN`: Assamese
65
+ * - `ur-IN`: Urdu
66
+ * - `ne-IN`: Nepali
67
+ * - `kok-IN`: Konkani
68
+ * - `ks-IN`: Kashmiri
69
+ * - `sd-IN`: Sindhi
70
+ * - `sa-IN`: Sanskrit
71
+ * - `sat-IN`: Santali
72
+ * - `mni-IN`: Manipuri
73
+ * - `brx-IN`: Bodo
74
+ * - `mai-IN`: Maithili
75
+ * - `doi-IN`: Dogri
62
76
  */
63
77
  language_code?: SarvamAI.SpeechToTextLanguage;
64
78
  /** Input Audio codec/format of the input file. PCM files are supported only at 16kHz sample rate. */
@@ -1,7 +1,8 @@
1
1
  /**
2
2
  * Specifies the language of the input audio in BCP-47 format.
3
3
  *
4
- * **Available Options:**
4
+ * **Available Options (saarika:v2.5):**
5
+ * - `unknown` (default): Use when the language is not known; the API will auto-detect.
5
6
  * - `hi-IN`: Hindi
6
7
  * - `bn-IN`: Bengali
7
8
  * - `gu-IN`: Gujarati
@@ -13,8 +14,23 @@
13
14
  * - `ta-IN`: Tamil
14
15
  * - `te-IN`: Telugu
15
16
  * - `en-IN`: English
17
+ *
18
+ * **Additional Options (saaras:v3 only):**
19
+ * - `as-IN`: Assamese
20
+ * - `ur-IN`: Urdu
21
+ * - `ne-IN`: Nepali
22
+ * - `kok-IN`: Konkani
23
+ * - `ks-IN`: Kashmiri
24
+ * - `sd-IN`: Sindhi
25
+ * - `sa-IN`: Sanskrit
26
+ * - `sat-IN`: Santali
27
+ * - `mni-IN`: Manipuri
28
+ * - `brx-IN`: Bodo
29
+ * - `mai-IN`: Maithili
30
+ * - `doi-IN`: Dogri
16
31
  */
17
32
  export declare const SpeechToTextStreamingLanguageCode: {
33
+ readonly Unknown: "unknown";
18
34
  readonly EnIn: "en-IN";
19
35
  readonly HiIn: "hi-IN";
20
36
  readonly BnIn: "bn-IN";
@@ -26,5 +42,17 @@ export declare const SpeechToTextStreamingLanguageCode: {
26
42
  readonly PaIn: "pa-IN";
27
43
  readonly TaIn: "ta-IN";
28
44
  readonly TeIn: "te-IN";
45
+ readonly AsIn: "as-IN";
46
+ readonly UrIn: "ur-IN";
47
+ readonly NeIn: "ne-IN";
48
+ readonly KokIn: "kok-IN";
49
+ readonly KsIn: "ks-IN";
50
+ readonly SdIn: "sd-IN";
51
+ readonly SaIn: "sa-IN";
52
+ readonly SatIn: "sat-IN";
53
+ readonly MniIn: "mni-IN";
54
+ readonly BrxIn: "brx-IN";
55
+ readonly MaiIn: "mai-IN";
56
+ readonly DoiIn: "doi-IN";
29
57
  };
30
58
  export type SpeechToTextStreamingLanguageCode = (typeof SpeechToTextStreamingLanguageCode)[keyof typeof SpeechToTextStreamingLanguageCode];
@@ -5,7 +5,8 @@ exports.SpeechToTextStreamingLanguageCode = void 0;
5
5
  /**
6
6
  * Specifies the language of the input audio in BCP-47 format.
7
7
  *
8
- * **Available Options:**
8
+ * **Available Options (saarika:v2.5):**
9
+ * - `unknown` (default): Use when the language is not known; the API will auto-detect.
9
10
  * - `hi-IN`: Hindi
10
11
  * - `bn-IN`: Bengali
11
12
  * - `gu-IN`: Gujarati
@@ -17,8 +18,23 @@ exports.SpeechToTextStreamingLanguageCode = void 0;
17
18
  * - `ta-IN`: Tamil
18
19
  * - `te-IN`: Telugu
19
20
  * - `en-IN`: English
21
+ *
22
+ * **Additional Options (saaras:v3 only):**
23
+ * - `as-IN`: Assamese
24
+ * - `ur-IN`: Urdu
25
+ * - `ne-IN`: Nepali
26
+ * - `kok-IN`: Konkani
27
+ * - `ks-IN`: Kashmiri
28
+ * - `sd-IN`: Sindhi
29
+ * - `sa-IN`: Sanskrit
30
+ * - `sat-IN`: Santali
31
+ * - `mni-IN`: Manipuri
32
+ * - `brx-IN`: Bodo
33
+ * - `mai-IN`: Maithili
34
+ * - `doi-IN`: Dogri
20
35
  */
21
36
  exports.SpeechToTextStreamingLanguageCode = {
37
+ Unknown: "unknown",
22
38
  EnIn: "en-IN",
23
39
  HiIn: "hi-IN",
24
40
  BnIn: "bn-IN",
@@ -30,4 +46,16 @@ exports.SpeechToTextStreamingLanguageCode = {
30
46
  PaIn: "pa-IN",
31
47
  TaIn: "ta-IN",
32
48
  TeIn: "te-IN",
49
+ AsIn: "as-IN",
50
+ UrIn: "ur-IN",
51
+ NeIn: "ne-IN",
52
+ KokIn: "kok-IN",
53
+ KsIn: "ks-IN",
54
+ SdIn: "sd-IN",
55
+ SaIn: "sa-IN",
56
+ SatIn: "sat-IN",
57
+ MniIn: "mni-IN",
58
+ BrxIn: "brx-IN",
59
+ MaiIn: "mai-IN",
60
+ DoiIn: "doi-IN",
33
61
  };
@@ -1,10 +1,10 @@
1
1
  /**
2
2
  * Text to speech model to use.
3
- * - **bulbul:v3**: Latest model with improved quality, 30+ voices (no pitch/loudness)
4
- * - **bulbul:v2** (default): Legacy TTS model with pitch/loudness/pace support
3
+ * - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
4
+ * - **bulbul:v3-beta**: Advanced model with temperature control (no pitch/loudness)
5
5
  */
6
6
  export declare const TextToSpeechStreamingModel: {
7
- readonly BulbulV3: "bulbul:v3";
8
7
  readonly BulbulV2: "bulbul:v2";
8
+ readonly BulbulV3Beta: "bulbul:v3-beta";
9
9
  };
10
10
  export type TextToSpeechStreamingModel = (typeof TextToSpeechStreamingModel)[keyof typeof TextToSpeechStreamingModel];
@@ -4,10 +4,10 @@ Object.defineProperty(exports, "__esModule", { value: true });
4
4
  exports.TextToSpeechStreamingModel = void 0;
5
5
  /**
6
6
  * Text to speech model to use.
7
- * - **bulbul:v3**: Latest model with improved quality, 30+ voices (no pitch/loudness)
8
- * - **bulbul:v2** (default): Legacy TTS model with pitch/loudness/pace support
7
+ * - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
8
+ * - **bulbul:v3-beta**: Advanced model with temperature control (no pitch/loudness)
9
9
  */
10
10
  exports.TextToSpeechStreamingModel = {
11
- BulbulV3: "bulbul:v3",
12
11
  BulbulV2: "bulbul:v2",
12
+ BulbulV3Beta: "bulbul:v3-beta",
13
13
  };
@@ -5,8 +5,8 @@
5
5
  * will be automatically flushed and processed before applying the new configuration.
6
6
  *
7
7
  * **Model-Specific Notes:**
8
- * - **bulbul:v3:** Latest model with improved quality, 30+ voices, pace (0.5-2.0), and temperature control. Does NOT support pitch or loudness. Default sample rate: 24000 Hz.
9
- * - **bulbul:v2:** Legacy model. Supports pitch, loudness, pace (0.3-3.0). Default sample rate: 22050 Hz.
8
+ * - **bulbul:v2:** Supports pitch, loudness, pace (0.3-3.0). Default sample rate: 22050 Hz.
9
+ * - **bulbul:v3-beta:** Does NOT support pitch/loudness. Pace range: 0.5-2.0. Supports temperature. Default sample rate: 24000 Hz.
10
10
  */
11
11
  export interface ConfigureConnection {
12
12
  type: "config";
@@ -16,18 +16,18 @@ export declare namespace ConfigureConnection {
16
16
  interface Data {
17
17
  /**
18
18
  * Specifies the model to use for text-to-speech conversion.
19
- * - **bulbul:v3**: Latest model with improved quality, 30+ voices (no pitch/loudness)
20
- * - **bulbul:v2** (default): Legacy TTS model with pitch/loudness/pace support
19
+ * - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
20
+ * - **bulbul:v3-beta**: Advanced model with temperature control (no pitch/loudness)
21
21
  */
22
22
  model?: Data.Model;
23
- /** The language code in BCP-47 format. Primarily used by the pre-TTS text normalization model to handle language-specific text processing such as number formatting, abbreviations, and special characters. */
23
+ /** The language of the text in BCP-47 format */
24
24
  target_language_code: Data.TargetLanguageCode;
25
25
  /**
26
26
  * The speaker voice to be used for the output audio.
27
27
  *
28
28
  * **Model Compatibility:**
29
- * - **bulbul:v3:** shubh (default), aditya, ritu, priya, neha, rahul, pooja, rohan, simran, kavya, amit, dev, ishita, shreya, ratan, varun, manan, sumit, roopa, kabir, aayan, ashutosh, advait, amelia, sophia, anand, tanya, tarun, sunny, mani, gokul, vijay, shruti, suhani, mohit, kavitha, rehan, soham, rupali
30
29
  * - **bulbul:v2:** anushka (default), abhilash, manisha, vidya, arya, karun, hitesh
30
+ * - **bulbul:v3-beta:** aditya (default), ritu, priya, neha, rahul, pooja, rohan, simran, kavya, amit, dev, ishita, shreya, ratan, varun, manan, sumit, roopa, kabir, aayan, shubh, ashutosh, advait, amelia, sophia
31
31
  *
32
32
  * **Note:** Speaker selection must match the chosen model version.
33
33
  */
@@ -37,7 +37,7 @@ export declare namespace ConfigureConnection {
37
37
  * while higher values make it sharper. The suitable range is between -0.75
38
38
  * and 0.75. Default is 0.0.
39
39
  *
40
- * **Note:** NOT supported for bulbul:v3. Will be ignored if provided.
40
+ * **Note:** NOT supported for bulbul:v3-beta. Will be ignored if provided.
41
41
  */
42
42
  pitch?: number;
43
43
  /**
@@ -45,8 +45,8 @@ export declare namespace ConfigureConnection {
45
45
  * while higher values make it faster. Default is 1.0.
46
46
  *
47
47
  * **Model-specific ranges:**
48
- * - **bulbul:v3:** 0.5 to 2.0
49
48
  * - **bulbul:v2:** 0.3 to 3.0
49
+ * - **bulbul:v3-beta:** 0.5 to 2.0
50
50
  */
51
51
  pace?: number;
52
52
  /**
@@ -54,7 +54,7 @@ export declare namespace ConfigureConnection {
54
54
  * while higher values make it louder. The suitable range is between 0.3
55
55
  * and 3.0. Default is 1.0.
56
56
  *
57
- * **Note:** NOT supported for bulbul:v3. Will be ignored if provided.
57
+ * **Note:** NOT supported for bulbul:v3-beta. Will be ignored if provided.
58
58
  */
59
59
  loudness?: number;
60
60
  /**
@@ -62,16 +62,16 @@ export declare namespace ConfigureConnection {
62
62
  * focused and deterministic, while higher values make it more random.
63
63
  * The suitable range is between 0.01 and 1.0. Default is 0.6.
64
64
  *
65
- * **Note:** Only supported for bulbul:v3. Will be ignored for bulbul:v2.
65
+ * **Note:** Only supported for bulbul:v3-beta. Will be ignored for bulbul:v2.
66
66
  */
67
67
  temperature?: number;
68
68
  /**
69
- * Specifies the sample rate of the output audio. For streaming, supported values are
69
+ * Specifies the sample rate of the output audio. Supported values are
70
70
  * 8000, 16000, 22050, 24000 Hz.
71
71
  *
72
- * **Note:** Higher sample rates (32000, 44100, 48000 Hz) are only available with bulbul:v3 via the REST API, not in streaming mode.
73
- *
74
- * **Default:** 24000 Hz
72
+ * **Model-specific defaults:**
73
+ * - **bulbul:v2:** 22050 Hz
74
+ * - **bulbul:v3-beta:** 24000 Hz
75
75
  */
76
76
  speech_sample_rate?: number;
77
77
  /**
@@ -79,7 +79,9 @@ export declare namespace ConfigureConnection {
79
79
  * (e.g., numbers, dates) is performed. Set to true for better handling
80
80
  * of mixed-language text.
81
81
  *
82
- * **Note:** Only supported for bulbul:v2. Not applicable for bulbul:v3.
82
+ * **Model-specific defaults:**
83
+ * - **bulbul:v2:** false (optional)
84
+ * - **bulbul:v3-beta:** Always enabled (cannot be disabled)
83
85
  */
84
86
  enable_preprocessing?: boolean;
85
87
  /** Audio codec (currently supports MP3 only, optimized for real-time playback) */
@@ -94,15 +96,15 @@ export declare namespace ConfigureConnection {
94
96
  namespace Data {
95
97
  /**
96
98
  * Specifies the model to use for text-to-speech conversion.
97
- * - **bulbul:v3**: Latest model with improved quality, 30+ voices (no pitch/loudness)
98
- * - **bulbul:v2** (default): Legacy TTS model with pitch/loudness/pace support
99
+ * - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
100
+ * - **bulbul:v3-beta**: Advanced model with temperature control (no pitch/loudness)
99
101
  */
100
102
  const Model: {
101
- readonly BulbulV3: "bulbul:v3";
102
103
  readonly BulbulV2: "bulbul:v2";
104
+ readonly BulbulV3Beta: "bulbul:v3-beta";
103
105
  };
104
106
  type Model = (typeof Model)[keyof typeof Model];
105
- /** The language code in BCP-47 format. Primarily used by the pre-TTS text normalization model to handle language-specific text processing such as number formatting, abbreviations, and special characters. */
107
+ /** The language of the text in BCP-47 format */
106
108
  const TargetLanguageCode: {
107
109
  readonly BnIn: "bn-IN";
108
110
  readonly EnIn: "en-IN";
@@ -121,8 +123,8 @@ export declare namespace ConfigureConnection {
121
123
  * The speaker voice to be used for the output audio.
122
124
  *
123
125
  * **Model Compatibility:**
124
- * - **bulbul:v3:** shubh (default), aditya, ritu, priya, neha, rahul, pooja, rohan, simran, kavya, amit, dev, ishita, shreya, ratan, varun, manan, sumit, roopa, kabir, aayan, ashutosh, advait, amelia, sophia, anand, tanya, tarun, sunny, mani, gokul, vijay, shruti, suhani, mohit, kavitha, rehan, soham, rupali
125
126
  * - **bulbul:v2:** anushka (default), abhilash, manisha, vidya, arya, karun, hitesh
127
+ * - **bulbul:v3-beta:** aditya (default), ritu, priya, neha, rahul, pooja, rohan, simran, kavya, amit, dev, ishita, shreya, ratan, varun, manan, sumit, roopa, kabir, aayan, shubh, ashutosh, advait, amelia, sophia
126
128
  *
127
129
  * **Note:** Speaker selection must match the chosen model version.
128
130
  */
@@ -159,20 +161,6 @@ export declare namespace ConfigureConnection {
159
161
  readonly Advait: "advait";
160
162
  readonly Amelia: "amelia";
161
163
  readonly Sophia: "sophia";
162
- readonly Anand: "anand";
163
- readonly Tanya: "tanya";
164
- readonly Tarun: "tarun";
165
- readonly Sunny: "sunny";
166
- readonly Mani: "mani";
167
- readonly Gokul: "gokul";
168
- readonly Vijay: "vijay";
169
- readonly Shruti: "shruti";
170
- readonly Suhani: "suhani";
171
- readonly Mohit: "mohit";
172
- readonly Kavitha: "kavitha";
173
- readonly Rehan: "rehan";
174
- readonly Soham: "soham";
175
- readonly Rupali: "rupali";
176
164
  };
177
165
  type Speaker = (typeof Speaker)[keyof typeof Speaker];
178
166
  /** Audio codec (currently supports MP3 only, optimized for real-time playback) */
@@ -8,14 +8,14 @@ var ConfigureConnection;
8
8
  (function (Data) {
9
9
  /**
10
10
  * Specifies the model to use for text-to-speech conversion.
11
- * - **bulbul:v3**: Latest model with improved quality, 30+ voices (no pitch/loudness)
12
- * - **bulbul:v2** (default): Legacy TTS model with pitch/loudness/pace support
11
+ * - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
12
+ * - **bulbul:v3-beta**: Advanced model with temperature control (no pitch/loudness)
13
13
  */
14
14
  Data.Model = {
15
- BulbulV3: "bulbul:v3",
16
15
  BulbulV2: "bulbul:v2",
16
+ BulbulV3Beta: "bulbul:v3-beta",
17
17
  };
18
- /** The language code in BCP-47 format. Primarily used by the pre-TTS text normalization model to handle language-specific text processing such as number formatting, abbreviations, and special characters. */
18
+ /** The language of the text in BCP-47 format */
19
19
  Data.TargetLanguageCode = {
20
20
  BnIn: "bn-IN",
21
21
  EnIn: "en-IN",
@@ -33,8 +33,8 @@ var ConfigureConnection;
33
33
  * The speaker voice to be used for the output audio.
34
34
  *
35
35
  * **Model Compatibility:**
36
- * - **bulbul:v3:** shubh (default), aditya, ritu, priya, neha, rahul, pooja, rohan, simran, kavya, amit, dev, ishita, shreya, ratan, varun, manan, sumit, roopa, kabir, aayan, ashutosh, advait, amelia, sophia, anand, tanya, tarun, sunny, mani, gokul, vijay, shruti, suhani, mohit, kavitha, rehan, soham, rupali
37
36
  * - **bulbul:v2:** anushka (default), abhilash, manisha, vidya, arya, karun, hitesh
37
+ * - **bulbul:v3-beta:** aditya (default), ritu, priya, neha, rahul, pooja, rohan, simran, kavya, amit, dev, ishita, shreya, ratan, varun, manan, sumit, roopa, kabir, aayan, shubh, ashutosh, advait, amelia, sophia
38
38
  *
39
39
  * **Note:** Speaker selection must match the chosen model version.
40
40
  */
@@ -71,20 +71,6 @@ var ConfigureConnection;
71
71
  Advait: "advait",
72
72
  Amelia: "amelia",
73
73
  Sophia: "sophia",
74
- Anand: "anand",
75
- Tanya: "tanya",
76
- Tarun: "tarun",
77
- Sunny: "sunny",
78
- Mani: "mani",
79
- Gokul: "gokul",
80
- Vijay: "vijay",
81
- Shruti: "shruti",
82
- Suhani: "suhani",
83
- Mohit: "mohit",
84
- Kavitha: "kavitha",
85
- Rehan: "rehan",
86
- Soham: "soham",
87
- Rupali: "rupali",
88
74
  };
89
75
  /** Audio codec (currently supports MP3 only, optimized for real-time playback) */
90
76
  Data.OutputAudioCodec = {
@@ -16,8 +16,22 @@ export interface SpeechToTextJobParameters {
16
16
  * - `te-IN`: Telugu
17
17
  * - `en-IN`: English
18
18
  * - `gu-IN`: Gujarati
19
+ *
20
+ * **Additional Options (saaras:v3 only):**
21
+ * - `as-IN`: Assamese
22
+ * - `ur-IN`: Urdu
23
+ * - `ne-IN`: Nepali
24
+ * - `kok-IN`: Konkani
25
+ * - `ks-IN`: Kashmiri
26
+ * - `sd-IN`: Sindhi
27
+ * - `sa-IN`: Sanskrit
28
+ * - `sat-IN`: Santali
29
+ * - `mni-IN`: Manipuri
30
+ * - `brx-IN`: Bodo
31
+ * - `mai-IN`: Maithili
32
+ * - `doi-IN`: Dogri
19
33
  */
20
- language_code?: SarvamAI.SpeechToTextTranslateLanguage;
34
+ language_code?: SarvamAI.SpeechToTextLanguage;
21
35
  /**
22
36
  * Model to be used for speech to text.
23
37
  *
@@ -1,3 +1,10 @@
1
+ /**
2
+ * Languages supported for Speech-to-Text.
3
+ *
4
+ * **saarika:v2.5 supports (12 languages):** unknown, hi-IN, bn-IN, kn-IN, ml-IN, mr-IN, od-IN, pa-IN, ta-IN, te-IN, en-IN, gu-IN
5
+ *
6
+ * **saaras:v3 supports all 23 languages** including: as-IN, ur-IN, ne-IN, kok-IN, ks-IN, sd-IN, sa-IN, sat-IN, mni-IN, brx-IN, mai-IN, doi-IN
7
+ */
1
8
  export declare const SpeechToTextLanguage: {
2
9
  readonly Unknown: "unknown";
3
10
  readonly HiIn: "hi-IN";
@@ -11,5 +18,17 @@ export declare const SpeechToTextLanguage: {
11
18
  readonly TeIn: "te-IN";
12
19
  readonly EnIn: "en-IN";
13
20
  readonly GuIn: "gu-IN";
21
+ readonly AsIn: "as-IN";
22
+ readonly UrIn: "ur-IN";
23
+ readonly NeIn: "ne-IN";
24
+ readonly KokIn: "kok-IN";
25
+ readonly KsIn: "ks-IN";
26
+ readonly SdIn: "sd-IN";
27
+ readonly SaIn: "sa-IN";
28
+ readonly SatIn: "sat-IN";
29
+ readonly MniIn: "mni-IN";
30
+ readonly BrxIn: "brx-IN";
31
+ readonly MaiIn: "mai-IN";
32
+ readonly DoiIn: "doi-IN";
14
33
  };
15
34
  export type SpeechToTextLanguage = (typeof SpeechToTextLanguage)[keyof typeof SpeechToTextLanguage];
@@ -2,6 +2,13 @@
2
2
  // This file was auto-generated by Fern from our API Definition.
3
3
  Object.defineProperty(exports, "__esModule", { value: true });
4
4
  exports.SpeechToTextLanguage = void 0;
5
+ /**
6
+ * Languages supported for Speech-to-Text.
7
+ *
8
+ * **saarika:v2.5 supports (12 languages):** unknown, hi-IN, bn-IN, kn-IN, ml-IN, mr-IN, od-IN, pa-IN, ta-IN, te-IN, en-IN, gu-IN
9
+ *
10
+ * **saaras:v3 supports all 23 languages** including: as-IN, ur-IN, ne-IN, kok-IN, ks-IN, sd-IN, sa-IN, sat-IN, mni-IN, brx-IN, mai-IN, doi-IN
11
+ */
5
12
  exports.SpeechToTextLanguage = {
6
13
  Unknown: "unknown",
7
14
  HiIn: "hi-IN",
@@ -15,4 +22,16 @@ exports.SpeechToTextLanguage = {
15
22
  TeIn: "te-IN",
16
23
  EnIn: "en-IN",
17
24
  GuIn: "gu-IN",
25
+ AsIn: "as-IN",
26
+ UrIn: "ur-IN",
27
+ NeIn: "ne-IN",
28
+ KokIn: "kok-IN",
29
+ KsIn: "ks-IN",
30
+ SdIn: "sd-IN",
31
+ SaIn: "sa-IN",
32
+ SatIn: "sat-IN",
33
+ MniIn: "mni-IN",
34
+ BrxIn: "brx-IN",
35
+ MaiIn: "mai-IN",
36
+ DoiIn: "doi-IN",
18
37
  };
@@ -9,4 +9,17 @@ export interface SpeechToTextResponse {
9
9
  diarized_transcript?: SarvamAI.DiarizedTranscript;
10
10
  /** This will return the BCP-47 code of language spoken in the input. If multiple languages are detected, this will return language code of most predominant spoken language. If no language is detected, this will be null */
11
11
  language_code?: string;
12
+ /**
13
+ * Float value (0.0 to 1.0) indicating the probability of the detected language being correct. Higher values indicate higher confidence.
14
+ *
15
+ * **When it returns a value:**
16
+ * - When `language_code` is not provided in the request
17
+ * - When `language_code` is set to `unknown`
18
+ *
19
+ * **When it returns null:**
20
+ * - When a specific `language_code` is provided (language detection is skipped)
21
+ *
22
+ * The parameter is always present in the response.
23
+ */
24
+ language_probability?: number;
12
25
  }
@@ -10,5 +10,18 @@ export interface SpeechToTextTranscriptionData {
10
10
  diarized_transcript?: Record<string, unknown>;
11
11
  /** BCP-47 code of detected language */
12
12
  language_code?: string;
13
+ /**
14
+ * Float value (0.0 to 1.0) indicating the probability of the detected language being correct. Higher values indicate higher confidence.
15
+ *
16
+ * **When it returns a value:**
17
+ * - When `language_code` is not provided in the request
18
+ * - When `language_code` is set to `unknown`
19
+ *
20
+ * **When it returns null:**
21
+ * - When a specific `language_code` is provided (language detection is skipped)
22
+ *
23
+ * The parameter is always present in the response.
24
+ */
25
+ language_probability?: number;
13
26
  metrics: SarvamAI.TranscriptionMetrics;
14
27
  }
@@ -1,3 +1,10 @@
1
+ /**
2
+ * Languages supported for Speech-to-Text-Translate (detected source language).
3
+ *
4
+ * **saaras:v2.5 supports (11 languages):** hi-IN, bn-IN, kn-IN, ml-IN, mr-IN, od-IN, pa-IN, ta-IN, te-IN, gu-IN, en-IN
5
+ *
6
+ * **All 22 languages available** including: as-IN, ur-IN, ne-IN, kok-IN, ks-IN, sd-IN, sa-IN, sat-IN, mni-IN, brx-IN, mai-IN, doi-IN
7
+ */
1
8
  export declare const SpeechToTextTranslateLanguage: {
2
9
  readonly HiIn: "hi-IN";
3
10
  readonly BnIn: "bn-IN";
@@ -10,5 +17,17 @@ export declare const SpeechToTextTranslateLanguage: {
10
17
  readonly TeIn: "te-IN";
11
18
  readonly GuIn: "gu-IN";
12
19
  readonly EnIn: "en-IN";
20
+ readonly AsIn: "as-IN";
21
+ readonly UrIn: "ur-IN";
22
+ readonly NeIn: "ne-IN";
23
+ readonly KokIn: "kok-IN";
24
+ readonly KsIn: "ks-IN";
25
+ readonly SdIn: "sd-IN";
26
+ readonly SaIn: "sa-IN";
27
+ readonly SatIn: "sat-IN";
28
+ readonly MniIn: "mni-IN";
29
+ readonly BrxIn: "brx-IN";
30
+ readonly MaiIn: "mai-IN";
31
+ readonly DoiIn: "doi-IN";
13
32
  };
14
33
  export type SpeechToTextTranslateLanguage = (typeof SpeechToTextTranslateLanguage)[keyof typeof SpeechToTextTranslateLanguage];
@@ -2,6 +2,13 @@
2
2
  // This file was auto-generated by Fern from our API Definition.
3
3
  Object.defineProperty(exports, "__esModule", { value: true });
4
4
  exports.SpeechToTextTranslateLanguage = void 0;
5
+ /**
6
+ * Languages supported for Speech-to-Text-Translate (detected source language).
7
+ *
8
+ * **saaras:v2.5 supports (11 languages):** hi-IN, bn-IN, kn-IN, ml-IN, mr-IN, od-IN, pa-IN, ta-IN, te-IN, gu-IN, en-IN
9
+ *
10
+ * **All 22 languages available** including: as-IN, ur-IN, ne-IN, kok-IN, ks-IN, sd-IN, sa-IN, sat-IN, mni-IN, brx-IN, mai-IN, doi-IN
11
+ */
5
12
  exports.SpeechToTextTranslateLanguage = {
6
13
  HiIn: "hi-IN",
7
14
  BnIn: "bn-IN",
@@ -14,4 +21,16 @@ exports.SpeechToTextTranslateLanguage = {
14
21
  TeIn: "te-IN",
15
22
  GuIn: "gu-IN",
16
23
  EnIn: "en-IN",
24
+ AsIn: "as-IN",
25
+ UrIn: "ur-IN",
26
+ NeIn: "ne-IN",
27
+ KokIn: "kok-IN",
28
+ KsIn: "ks-IN",
29
+ SdIn: "sd-IN",
30
+ SaIn: "sa-IN",
31
+ SatIn: "sat-IN",
32
+ MniIn: "mni-IN",
33
+ BrxIn: "brx-IN",
34
+ MaiIn: "mai-IN",
35
+ DoiIn: "doi-IN",
17
36
  };
@@ -7,4 +7,17 @@ export interface SpeechToTextTranslateResponse {
7
7
  language_code?: SarvamAI.SpeechToTextTranslateLanguage;
8
8
  /** Diarized transcript of the provided speech */
9
9
  diarized_transcript?: SarvamAI.DiarizedTranscript;
10
+ /**
11
+ * Float value (0.0 to 1.0) indicating the probability of the detected language being correct. Higher values indicate higher confidence.
12
+ *
13
+ * **When it returns a value:**
14
+ * - When `language_code` is not provided in the request
15
+ * - When `language_code` is set to `unknown`
16
+ *
17
+ * **When it returns null:**
18
+ * - When a specific `language_code` is provided (language detection is skipped)
19
+ *
20
+ * The parameter is always present in the response.
21
+ */
22
+ language_probability?: number;
10
23
  }
@@ -6,5 +6,18 @@ export interface SpeechToTextTranslateTranscriptionData {
6
6
  transcript: string;
7
7
  /** BCP-47 code of detected source language (null when language detection is in progress) */
8
8
  language_code?: string;
9
+ /**
10
+ * Float value (0.0 to 1.0) indicating the probability of the detected language being correct. Higher values indicate higher confidence.
11
+ *
12
+ * **When it returns a value:**
13
+ * - When `language_code` is not provided in the request
14
+ * - When `language_code` is set to `unknown`
15
+ *
16
+ * **When it returns null:**
17
+ * - When a specific `language_code` is provided (language detection is skipped)
18
+ *
19
+ * The parameter is always present in the response.
20
+ */
21
+ language_probability?: number;
9
22
  metrics: SarvamAI.TranscriptionMetrics;
10
23
  }
@@ -1,5 +1,5 @@
1
1
  export declare const TextToSpeechModel: {
2
- readonly BulbulV3: "bulbul:v3";
3
2
  readonly BulbulV2: "bulbul:v2";
3
+ readonly BulbulV3: "bulbul:v3";
4
4
  };
5
5
  export type TextToSpeechModel = (typeof TextToSpeechModel)[keyof typeof TextToSpeechModel];
@@ -3,6 +3,6 @@
3
3
  Object.defineProperty(exports, "__esModule", { value: true });
4
4
  exports.TextToSpeechModel = void 0;
5
5
  exports.TextToSpeechModel = {
6
- BulbulV3: "bulbul:v3",
7
6
  BulbulV2: "bulbul:v2",
7
+ BulbulV3: "bulbul:v3",
8
8
  };
@@ -1 +1 @@
1
- export declare const SDK_VERSION = "1.1.3";
1
+ export declare const SDK_VERSION = "1.1.4";
@@ -1,4 +1,4 @@
1
1
  "use strict";
2
2
  Object.defineProperty(exports, "__esModule", { value: true });
3
3
  exports.SDK_VERSION = void 0;
4
- exports.SDK_VERSION = "1.1.3";
4
+ exports.SDK_VERSION = "1.1.4";
@@ -6,8 +6,8 @@ export function normalizeClientOptions(options) {
6
6
  const headers = mergeHeaders({
7
7
  "X-Fern-Language": "JavaScript",
8
8
  "X-Fern-SDK-Name": "sarvamai",
9
- "X-Fern-SDK-Version": "1.1.3",
10
- "User-Agent": "sarvamai/1.1.3",
9
+ "X-Fern-SDK-Version": "1.1.4",
10
+ "User-Agent": "sarvamai/1.1.4",
11
11
  "X-Fern-Runtime": core.RUNTIME.type,
12
12
  "X-Fern-Runtime-Version": core.RUNTIME.version,
13
13
  }, options === null || options === void 0 ? void 0 : options.headers);
@@ -59,6 +59,20 @@ export interface SpeechToTextTranscriptionRequest {
59
59
  * - `te-IN`: Telugu
60
60
  * - `en-IN`: English
61
61
  * - `gu-IN`: Gujarati
62
+ *
63
+ * **Additional Options (saaras:v3 only):**
64
+ * - `as-IN`: Assamese
65
+ * - `ur-IN`: Urdu
66
+ * - `ne-IN`: Nepali
67
+ * - `kok-IN`: Konkani
68
+ * - `ks-IN`: Kashmiri
69
+ * - `sd-IN`: Sindhi
70
+ * - `sa-IN`: Sanskrit
71
+ * - `sat-IN`: Santali
72
+ * - `mni-IN`: Manipuri
73
+ * - `brx-IN`: Bodo
74
+ * - `mai-IN`: Maithili
75
+ * - `doi-IN`: Dogri
62
76
  */
63
77
  language_code?: SarvamAI.SpeechToTextLanguage;
64
78
  /** Input Audio codec/format of the input file. PCM files are supported only at 16kHz sample rate. */
@@ -1,7 +1,8 @@
1
1
  /**
2
2
  * Specifies the language of the input audio in BCP-47 format.
3
3
  *
4
- * **Available Options:**
4
+ * **Available Options (saarika:v2.5):**
5
+ * - `unknown` (default): Use when the language is not known; the API will auto-detect.
5
6
  * - `hi-IN`: Hindi
6
7
  * - `bn-IN`: Bengali
7
8
  * - `gu-IN`: Gujarati
@@ -13,8 +14,23 @@
13
14
  * - `ta-IN`: Tamil
14
15
  * - `te-IN`: Telugu
15
16
  * - `en-IN`: English
17
+ *
18
+ * **Additional Options (saaras:v3 only):**
19
+ * - `as-IN`: Assamese
20
+ * - `ur-IN`: Urdu
21
+ * - `ne-IN`: Nepali
22
+ * - `kok-IN`: Konkani
23
+ * - `ks-IN`: Kashmiri
24
+ * - `sd-IN`: Sindhi
25
+ * - `sa-IN`: Sanskrit
26
+ * - `sat-IN`: Santali
27
+ * - `mni-IN`: Manipuri
28
+ * - `brx-IN`: Bodo
29
+ * - `mai-IN`: Maithili
30
+ * - `doi-IN`: Dogri
16
31
  */
17
32
  export declare const SpeechToTextStreamingLanguageCode: {
33
+ readonly Unknown: "unknown";
18
34
  readonly EnIn: "en-IN";
19
35
  readonly HiIn: "hi-IN";
20
36
  readonly BnIn: "bn-IN";
@@ -26,5 +42,17 @@ export declare const SpeechToTextStreamingLanguageCode: {
26
42
  readonly PaIn: "pa-IN";
27
43
  readonly TaIn: "ta-IN";
28
44
  readonly TeIn: "te-IN";
45
+ readonly AsIn: "as-IN";
46
+ readonly UrIn: "ur-IN";
47
+ readonly NeIn: "ne-IN";
48
+ readonly KokIn: "kok-IN";
49
+ readonly KsIn: "ks-IN";
50
+ readonly SdIn: "sd-IN";
51
+ readonly SaIn: "sa-IN";
52
+ readonly SatIn: "sat-IN";
53
+ readonly MniIn: "mni-IN";
54
+ readonly BrxIn: "brx-IN";
55
+ readonly MaiIn: "mai-IN";
56
+ readonly DoiIn: "doi-IN";
29
57
  };
30
58
  export type SpeechToTextStreamingLanguageCode = (typeof SpeechToTextStreamingLanguageCode)[keyof typeof SpeechToTextStreamingLanguageCode];
@@ -2,7 +2,8 @@
2
2
  /**
3
3
  * Specifies the language of the input audio in BCP-47 format.
4
4
  *
5
- * **Available Options:**
5
+ * **Available Options (saarika:v2.5):**
6
+ * - `unknown` (default): Use when the language is not known; the API will auto-detect.
6
7
  * - `hi-IN`: Hindi
7
8
  * - `bn-IN`: Bengali
8
9
  * - `gu-IN`: Gujarati
@@ -14,8 +15,23 @@
14
15
  * - `ta-IN`: Tamil
15
16
  * - `te-IN`: Telugu
16
17
  * - `en-IN`: English
18
+ *
19
+ * **Additional Options (saaras:v3 only):**
20
+ * - `as-IN`: Assamese
21
+ * - `ur-IN`: Urdu
22
+ * - `ne-IN`: Nepali
23
+ * - `kok-IN`: Konkani
24
+ * - `ks-IN`: Kashmiri
25
+ * - `sd-IN`: Sindhi
26
+ * - `sa-IN`: Sanskrit
27
+ * - `sat-IN`: Santali
28
+ * - `mni-IN`: Manipuri
29
+ * - `brx-IN`: Bodo
30
+ * - `mai-IN`: Maithili
31
+ * - `doi-IN`: Dogri
17
32
  */
18
33
  export const SpeechToTextStreamingLanguageCode = {
34
+ Unknown: "unknown",
19
35
  EnIn: "en-IN",
20
36
  HiIn: "hi-IN",
21
37
  BnIn: "bn-IN",
@@ -27,4 +43,16 @@ export const SpeechToTextStreamingLanguageCode = {
27
43
  PaIn: "pa-IN",
28
44
  TaIn: "ta-IN",
29
45
  TeIn: "te-IN",
46
+ AsIn: "as-IN",
47
+ UrIn: "ur-IN",
48
+ NeIn: "ne-IN",
49
+ KokIn: "kok-IN",
50
+ KsIn: "ks-IN",
51
+ SdIn: "sd-IN",
52
+ SaIn: "sa-IN",
53
+ SatIn: "sat-IN",
54
+ MniIn: "mni-IN",
55
+ BrxIn: "brx-IN",
56
+ MaiIn: "mai-IN",
57
+ DoiIn: "doi-IN",
30
58
  };
@@ -1,10 +1,10 @@
1
1
  /**
2
2
  * Text to speech model to use.
3
- * - **bulbul:v3**: Latest model with improved quality, 30+ voices (no pitch/loudness)
4
- * - **bulbul:v2** (default): Legacy TTS model with pitch/loudness/pace support
3
+ * - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
4
+ * - **bulbul:v3-beta**: Advanced model with temperature control (no pitch/loudness)
5
5
  */
6
6
  export declare const TextToSpeechStreamingModel: {
7
- readonly BulbulV3: "bulbul:v3";
8
7
  readonly BulbulV2: "bulbul:v2";
8
+ readonly BulbulV3Beta: "bulbul:v3-beta";
9
9
  };
10
10
  export type TextToSpeechStreamingModel = (typeof TextToSpeechStreamingModel)[keyof typeof TextToSpeechStreamingModel];
@@ -1,10 +1,10 @@
1
1
  // This file was auto-generated by Fern from our API Definition.
2
2
  /**
3
3
  * Text to speech model to use.
4
- * - **bulbul:v3**: Latest model with improved quality, 30+ voices (no pitch/loudness)
5
- * - **bulbul:v2** (default): Legacy TTS model with pitch/loudness/pace support
4
+ * - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
5
+ * - **bulbul:v3-beta**: Advanced model with temperature control (no pitch/loudness)
6
6
  */
7
7
  export const TextToSpeechStreamingModel = {
8
- BulbulV3: "bulbul:v3",
9
8
  BulbulV2: "bulbul:v2",
9
+ BulbulV3Beta: "bulbul:v3-beta",
10
10
  };
@@ -5,8 +5,8 @@
5
5
  * will be automatically flushed and processed before applying the new configuration.
6
6
  *
7
7
  * **Model-Specific Notes:**
8
- * - **bulbul:v3:** Latest model with improved quality, 30+ voices, pace (0.5-2.0), and temperature control. Does NOT support pitch or loudness. Default sample rate: 24000 Hz.
9
- * - **bulbul:v2:** Legacy model. Supports pitch, loudness, pace (0.3-3.0). Default sample rate: 22050 Hz.
8
+ * - **bulbul:v2:** Supports pitch, loudness, pace (0.3-3.0). Default sample rate: 22050 Hz.
9
+ * - **bulbul:v3-beta:** Does NOT support pitch/loudness. Pace range: 0.5-2.0. Supports temperature. Default sample rate: 24000 Hz.
10
10
  */
11
11
  export interface ConfigureConnection {
12
12
  type: "config";
@@ -16,18 +16,18 @@ export declare namespace ConfigureConnection {
16
16
  interface Data {
17
17
  /**
18
18
  * Specifies the model to use for text-to-speech conversion.
19
- * - **bulbul:v3**: Latest model with improved quality, 30+ voices (no pitch/loudness)
20
- * - **bulbul:v2** (default): Legacy TTS model with pitch/loudness/pace support
19
+ * - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
20
+ * - **bulbul:v3-beta**: Advanced model with temperature control (no pitch/loudness)
21
21
  */
22
22
  model?: Data.Model;
23
- /** The language code in BCP-47 format. Primarily used by the pre-TTS text normalization model to handle language-specific text processing such as number formatting, abbreviations, and special characters. */
23
+ /** The language of the text in BCP-47 format */
24
24
  target_language_code: Data.TargetLanguageCode;
25
25
  /**
26
26
  * The speaker voice to be used for the output audio.
27
27
  *
28
28
  * **Model Compatibility:**
29
- * - **bulbul:v3:** shubh (default), aditya, ritu, priya, neha, rahul, pooja, rohan, simran, kavya, amit, dev, ishita, shreya, ratan, varun, manan, sumit, roopa, kabir, aayan, ashutosh, advait, amelia, sophia, anand, tanya, tarun, sunny, mani, gokul, vijay, shruti, suhani, mohit, kavitha, rehan, soham, rupali
30
29
  * - **bulbul:v2:** anushka (default), abhilash, manisha, vidya, arya, karun, hitesh
30
+ * - **bulbul:v3-beta:** aditya (default), ritu, priya, neha, rahul, pooja, rohan, simran, kavya, amit, dev, ishita, shreya, ratan, varun, manan, sumit, roopa, kabir, aayan, shubh, ashutosh, advait, amelia, sophia
31
31
  *
32
32
  * **Note:** Speaker selection must match the chosen model version.
33
33
  */
@@ -37,7 +37,7 @@ export declare namespace ConfigureConnection {
37
37
  * while higher values make it sharper. The suitable range is between -0.75
38
38
  * and 0.75. Default is 0.0.
39
39
  *
40
- * **Note:** NOT supported for bulbul:v3. Will be ignored if provided.
40
+ * **Note:** NOT supported for bulbul:v3-beta. Will be ignored if provided.
41
41
  */
42
42
  pitch?: number;
43
43
  /**
@@ -45,8 +45,8 @@ export declare namespace ConfigureConnection {
45
45
  * while higher values make it faster. Default is 1.0.
46
46
  *
47
47
  * **Model-specific ranges:**
48
- * - **bulbul:v3:** 0.5 to 2.0
49
48
  * - **bulbul:v2:** 0.3 to 3.0
49
+ * - **bulbul:v3-beta:** 0.5 to 2.0
50
50
  */
51
51
  pace?: number;
52
52
  /**
@@ -54,7 +54,7 @@ export declare namespace ConfigureConnection {
54
54
  * while higher values make it louder. The suitable range is between 0.3
55
55
  * and 3.0. Default is 1.0.
56
56
  *
57
- * **Note:** NOT supported for bulbul:v3. Will be ignored if provided.
57
+ * **Note:** NOT supported for bulbul:v3-beta. Will be ignored if provided.
58
58
  */
59
59
  loudness?: number;
60
60
  /**
@@ -62,16 +62,16 @@ export declare namespace ConfigureConnection {
62
62
  * focused and deterministic, while higher values make it more random.
63
63
  * The suitable range is between 0.01 and 1.0. Default is 0.6.
64
64
  *
65
- * **Note:** Only supported for bulbul:v3. Will be ignored for bulbul:v2.
65
+ * **Note:** Only supported for bulbul:v3-beta. Will be ignored for bulbul:v2.
66
66
  */
67
67
  temperature?: number;
68
68
  /**
69
- * Specifies the sample rate of the output audio. For streaming, supported values are
69
+ * Specifies the sample rate of the output audio. Supported values are
70
70
  * 8000, 16000, 22050, 24000 Hz.
71
71
  *
72
- * **Note:** Higher sample rates (32000, 44100, 48000 Hz) are only available with bulbul:v3 via the REST API, not in streaming mode.
73
- *
74
- * **Default:** 24000 Hz
72
+ * **Model-specific defaults:**
73
+ * - **bulbul:v2:** 22050 Hz
74
+ * - **bulbul:v3-beta:** 24000 Hz
75
75
  */
76
76
  speech_sample_rate?: number;
77
77
  /**
@@ -79,7 +79,9 @@ export declare namespace ConfigureConnection {
79
79
  * (e.g., numbers, dates) is performed. Set to true for better handling
80
80
  * of mixed-language text.
81
81
  *
82
- * **Note:** Only supported for bulbul:v2. Not applicable for bulbul:v3.
82
+ * **Model-specific defaults:**
83
+ * - **bulbul:v2:** false (optional)
84
+ * - **bulbul:v3-beta:** Always enabled (cannot be disabled)
83
85
  */
84
86
  enable_preprocessing?: boolean;
85
87
  /** Audio codec (currently supports MP3 only, optimized for real-time playback) */
@@ -94,15 +96,15 @@ export declare namespace ConfigureConnection {
94
96
  namespace Data {
95
97
  /**
96
98
  * Specifies the model to use for text-to-speech conversion.
97
- * - **bulbul:v3**: Latest model with improved quality, 30+ voices (no pitch/loudness)
98
- * - **bulbul:v2** (default): Legacy TTS model with pitch/loudness/pace support
99
+ * - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
100
+ * - **bulbul:v3-beta**: Advanced model with temperature control (no pitch/loudness)
99
101
  */
100
102
  const Model: {
101
- readonly BulbulV3: "bulbul:v3";
102
103
  readonly BulbulV2: "bulbul:v2";
104
+ readonly BulbulV3Beta: "bulbul:v3-beta";
103
105
  };
104
106
  type Model = (typeof Model)[keyof typeof Model];
105
- /** The language code in BCP-47 format. Primarily used by the pre-TTS text normalization model to handle language-specific text processing such as number formatting, abbreviations, and special characters. */
107
+ /** The language of the text in BCP-47 format */
106
108
  const TargetLanguageCode: {
107
109
  readonly BnIn: "bn-IN";
108
110
  readonly EnIn: "en-IN";
@@ -121,8 +123,8 @@ export declare namespace ConfigureConnection {
121
123
  * The speaker voice to be used for the output audio.
122
124
  *
123
125
  * **Model Compatibility:**
124
- * - **bulbul:v3:** shubh (default), aditya, ritu, priya, neha, rahul, pooja, rohan, simran, kavya, amit, dev, ishita, shreya, ratan, varun, manan, sumit, roopa, kabir, aayan, ashutosh, advait, amelia, sophia, anand, tanya, tarun, sunny, mani, gokul, vijay, shruti, suhani, mohit, kavitha, rehan, soham, rupali
125
126
  * - **bulbul:v2:** anushka (default), abhilash, manisha, vidya, arya, karun, hitesh
127
+ * - **bulbul:v3-beta:** aditya (default), ritu, priya, neha, rahul, pooja, rohan, simran, kavya, amit, dev, ishita, shreya, ratan, varun, manan, sumit, roopa, kabir, aayan, shubh, ashutosh, advait, amelia, sophia
126
128
  *
127
129
  * **Note:** Speaker selection must match the chosen model version.
128
130
  */
@@ -159,20 +161,6 @@ export declare namespace ConfigureConnection {
159
161
  readonly Advait: "advait";
160
162
  readonly Amelia: "amelia";
161
163
  readonly Sophia: "sophia";
162
- readonly Anand: "anand";
163
- readonly Tanya: "tanya";
164
- readonly Tarun: "tarun";
165
- readonly Sunny: "sunny";
166
- readonly Mani: "mani";
167
- readonly Gokul: "gokul";
168
- readonly Vijay: "vijay";
169
- readonly Shruti: "shruti";
170
- readonly Suhani: "suhani";
171
- readonly Mohit: "mohit";
172
- readonly Kavitha: "kavitha";
173
- readonly Rehan: "rehan";
174
- readonly Soham: "soham";
175
- readonly Rupali: "rupali";
176
164
  };
177
165
  type Speaker = (typeof Speaker)[keyof typeof Speaker];
178
166
  /** Audio codec (currently supports MP3 only, optimized for real-time playback) */
@@ -5,14 +5,14 @@ export var ConfigureConnection;
5
5
  (function (Data) {
6
6
  /**
7
7
  * Specifies the model to use for text-to-speech conversion.
8
- * - **bulbul:v3**: Latest model with improved quality, 30+ voices (no pitch/loudness)
9
- * - **bulbul:v2** (default): Legacy TTS model with pitch/loudness/pace support
8
+ * - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
9
+ * - **bulbul:v3-beta**: Advanced model with temperature control (no pitch/loudness)
10
10
  */
11
11
  Data.Model = {
12
- BulbulV3: "bulbul:v3",
13
12
  BulbulV2: "bulbul:v2",
13
+ BulbulV3Beta: "bulbul:v3-beta",
14
14
  };
15
- /** The language code in BCP-47 format. Primarily used by the pre-TTS text normalization model to handle language-specific text processing such as number formatting, abbreviations, and special characters. */
15
+ /** The language of the text in BCP-47 format */
16
16
  Data.TargetLanguageCode = {
17
17
  BnIn: "bn-IN",
18
18
  EnIn: "en-IN",
@@ -30,8 +30,8 @@ export var ConfigureConnection;
30
30
  * The speaker voice to be used for the output audio.
31
31
  *
32
32
  * **Model Compatibility:**
33
- * - **bulbul:v3:** shubh (default), aditya, ritu, priya, neha, rahul, pooja, rohan, simran, kavya, amit, dev, ishita, shreya, ratan, varun, manan, sumit, roopa, kabir, aayan, ashutosh, advait, amelia, sophia, anand, tanya, tarun, sunny, mani, gokul, vijay, shruti, suhani, mohit, kavitha, rehan, soham, rupali
34
33
  * - **bulbul:v2:** anushka (default), abhilash, manisha, vidya, arya, karun, hitesh
34
+ * - **bulbul:v3-beta:** aditya (default), ritu, priya, neha, rahul, pooja, rohan, simran, kavya, amit, dev, ishita, shreya, ratan, varun, manan, sumit, roopa, kabir, aayan, shubh, ashutosh, advait, amelia, sophia
35
35
  *
36
36
  * **Note:** Speaker selection must match the chosen model version.
37
37
  */
@@ -68,20 +68,6 @@ export var ConfigureConnection;
68
68
  Advait: "advait",
69
69
  Amelia: "amelia",
70
70
  Sophia: "sophia",
71
- Anand: "anand",
72
- Tanya: "tanya",
73
- Tarun: "tarun",
74
- Sunny: "sunny",
75
- Mani: "mani",
76
- Gokul: "gokul",
77
- Vijay: "vijay",
78
- Shruti: "shruti",
79
- Suhani: "suhani",
80
- Mohit: "mohit",
81
- Kavitha: "kavitha",
82
- Rehan: "rehan",
83
- Soham: "soham",
84
- Rupali: "rupali",
85
71
  };
86
72
  /** Audio codec (currently supports MP3 only, optimized for real-time playback) */
87
73
  Data.OutputAudioCodec = {
@@ -16,8 +16,22 @@ export interface SpeechToTextJobParameters {
16
16
  * - `te-IN`: Telugu
17
17
  * - `en-IN`: English
18
18
  * - `gu-IN`: Gujarati
19
+ *
20
+ * **Additional Options (saaras:v3 only):**
21
+ * - `as-IN`: Assamese
22
+ * - `ur-IN`: Urdu
23
+ * - `ne-IN`: Nepali
24
+ * - `kok-IN`: Konkani
25
+ * - `ks-IN`: Kashmiri
26
+ * - `sd-IN`: Sindhi
27
+ * - `sa-IN`: Sanskrit
28
+ * - `sat-IN`: Santali
29
+ * - `mni-IN`: Manipuri
30
+ * - `brx-IN`: Bodo
31
+ * - `mai-IN`: Maithili
32
+ * - `doi-IN`: Dogri
19
33
  */
20
- language_code?: SarvamAI.SpeechToTextTranslateLanguage;
34
+ language_code?: SarvamAI.SpeechToTextLanguage;
21
35
  /**
22
36
  * Model to be used for speech to text.
23
37
  *
@@ -1,3 +1,10 @@
1
+ /**
2
+ * Languages supported for Speech-to-Text.
3
+ *
4
+ * **saarika:v2.5 supports (12 languages):** unknown, hi-IN, bn-IN, kn-IN, ml-IN, mr-IN, od-IN, pa-IN, ta-IN, te-IN, en-IN, gu-IN
5
+ *
6
+ * **saaras:v3 supports all 23 languages** including: as-IN, ur-IN, ne-IN, kok-IN, ks-IN, sd-IN, sa-IN, sat-IN, mni-IN, brx-IN, mai-IN, doi-IN
7
+ */
1
8
  export declare const SpeechToTextLanguage: {
2
9
  readonly Unknown: "unknown";
3
10
  readonly HiIn: "hi-IN";
@@ -11,5 +18,17 @@ export declare const SpeechToTextLanguage: {
11
18
  readonly TeIn: "te-IN";
12
19
  readonly EnIn: "en-IN";
13
20
  readonly GuIn: "gu-IN";
21
+ readonly AsIn: "as-IN";
22
+ readonly UrIn: "ur-IN";
23
+ readonly NeIn: "ne-IN";
24
+ readonly KokIn: "kok-IN";
25
+ readonly KsIn: "ks-IN";
26
+ readonly SdIn: "sd-IN";
27
+ readonly SaIn: "sa-IN";
28
+ readonly SatIn: "sat-IN";
29
+ readonly MniIn: "mni-IN";
30
+ readonly BrxIn: "brx-IN";
31
+ readonly MaiIn: "mai-IN";
32
+ readonly DoiIn: "doi-IN";
14
33
  };
15
34
  export type SpeechToTextLanguage = (typeof SpeechToTextLanguage)[keyof typeof SpeechToTextLanguage];
@@ -1,4 +1,11 @@
1
1
  // This file was auto-generated by Fern from our API Definition.
2
+ /**
3
+ * Languages supported for Speech-to-Text.
4
+ *
5
+ * **saarika:v2.5 supports (12 languages):** unknown, hi-IN, bn-IN, kn-IN, ml-IN, mr-IN, od-IN, pa-IN, ta-IN, te-IN, en-IN, gu-IN
6
+ *
7
+ * **saaras:v3 supports all 23 languages** including: as-IN, ur-IN, ne-IN, kok-IN, ks-IN, sd-IN, sa-IN, sat-IN, mni-IN, brx-IN, mai-IN, doi-IN
8
+ */
2
9
  export const SpeechToTextLanguage = {
3
10
  Unknown: "unknown",
4
11
  HiIn: "hi-IN",
@@ -12,4 +19,16 @@ export const SpeechToTextLanguage = {
12
19
  TeIn: "te-IN",
13
20
  EnIn: "en-IN",
14
21
  GuIn: "gu-IN",
22
+ AsIn: "as-IN",
23
+ UrIn: "ur-IN",
24
+ NeIn: "ne-IN",
25
+ KokIn: "kok-IN",
26
+ KsIn: "ks-IN",
27
+ SdIn: "sd-IN",
28
+ SaIn: "sa-IN",
29
+ SatIn: "sat-IN",
30
+ MniIn: "mni-IN",
31
+ BrxIn: "brx-IN",
32
+ MaiIn: "mai-IN",
33
+ DoiIn: "doi-IN",
15
34
  };
@@ -9,4 +9,17 @@ export interface SpeechToTextResponse {
9
9
  diarized_transcript?: SarvamAI.DiarizedTranscript;
10
10
  /** This will return the BCP-47 code of language spoken in the input. If multiple languages are detected, this will return language code of most predominant spoken language. If no language is detected, this will be null */
11
11
  language_code?: string;
12
+ /**
13
+ * Float value (0.0 to 1.0) indicating the probability of the detected language being correct. Higher values indicate higher confidence.
14
+ *
15
+ * **When it returns a value:**
16
+ * - When `language_code` is not provided in the request
17
+ * - When `language_code` is set to `unknown`
18
+ *
19
+ * **When it returns null:**
20
+ * - When a specific `language_code` is provided (language detection is skipped)
21
+ *
22
+ * The parameter is always present in the response.
23
+ */
24
+ language_probability?: number;
12
25
  }
@@ -10,5 +10,18 @@ export interface SpeechToTextTranscriptionData {
10
10
  diarized_transcript?: Record<string, unknown>;
11
11
  /** BCP-47 code of detected language */
12
12
  language_code?: string;
13
+ /**
14
+ * Float value (0.0 to 1.0) indicating the probability of the detected language being correct. Higher values indicate higher confidence.
15
+ *
16
+ * **When it returns a value:**
17
+ * - When `language_code` is not provided in the request
18
+ * - When `language_code` is set to `unknown`
19
+ *
20
+ * **When it returns null:**
21
+ * - When a specific `language_code` is provided (language detection is skipped)
22
+ *
23
+ * The parameter is always present in the response.
24
+ */
25
+ language_probability?: number;
13
26
  metrics: SarvamAI.TranscriptionMetrics;
14
27
  }
@@ -1,3 +1,10 @@
1
+ /**
2
+ * Languages supported for Speech-to-Text-Translate (detected source language).
3
+ *
4
+ * **saaras:v2.5 supports (11 languages):** hi-IN, bn-IN, kn-IN, ml-IN, mr-IN, od-IN, pa-IN, ta-IN, te-IN, gu-IN, en-IN
5
+ *
6
+ * **All 22 languages available** including: as-IN, ur-IN, ne-IN, kok-IN, ks-IN, sd-IN, sa-IN, sat-IN, mni-IN, brx-IN, mai-IN, doi-IN
7
+ */
1
8
  export declare const SpeechToTextTranslateLanguage: {
2
9
  readonly HiIn: "hi-IN";
3
10
  readonly BnIn: "bn-IN";
@@ -10,5 +17,17 @@ export declare const SpeechToTextTranslateLanguage: {
10
17
  readonly TeIn: "te-IN";
11
18
  readonly GuIn: "gu-IN";
12
19
  readonly EnIn: "en-IN";
20
+ readonly AsIn: "as-IN";
21
+ readonly UrIn: "ur-IN";
22
+ readonly NeIn: "ne-IN";
23
+ readonly KokIn: "kok-IN";
24
+ readonly KsIn: "ks-IN";
25
+ readonly SdIn: "sd-IN";
26
+ readonly SaIn: "sa-IN";
27
+ readonly SatIn: "sat-IN";
28
+ readonly MniIn: "mni-IN";
29
+ readonly BrxIn: "brx-IN";
30
+ readonly MaiIn: "mai-IN";
31
+ readonly DoiIn: "doi-IN";
13
32
  };
14
33
  export type SpeechToTextTranslateLanguage = (typeof SpeechToTextTranslateLanguage)[keyof typeof SpeechToTextTranslateLanguage];
@@ -1,4 +1,11 @@
1
1
  // This file was auto-generated by Fern from our API Definition.
2
+ /**
3
+ * Languages supported for Speech-to-Text-Translate (detected source language).
4
+ *
5
+ * **saaras:v2.5 supports (11 languages):** hi-IN, bn-IN, kn-IN, ml-IN, mr-IN, od-IN, pa-IN, ta-IN, te-IN, gu-IN, en-IN
6
+ *
7
+ * **All 22 languages available** including: as-IN, ur-IN, ne-IN, kok-IN, ks-IN, sd-IN, sa-IN, sat-IN, mni-IN, brx-IN, mai-IN, doi-IN
8
+ */
2
9
  export const SpeechToTextTranslateLanguage = {
3
10
  HiIn: "hi-IN",
4
11
  BnIn: "bn-IN",
@@ -11,4 +18,16 @@ export const SpeechToTextTranslateLanguage = {
11
18
  TeIn: "te-IN",
12
19
  GuIn: "gu-IN",
13
20
  EnIn: "en-IN",
21
+ AsIn: "as-IN",
22
+ UrIn: "ur-IN",
23
+ NeIn: "ne-IN",
24
+ KokIn: "kok-IN",
25
+ KsIn: "ks-IN",
26
+ SdIn: "sd-IN",
27
+ SaIn: "sa-IN",
28
+ SatIn: "sat-IN",
29
+ MniIn: "mni-IN",
30
+ BrxIn: "brx-IN",
31
+ MaiIn: "mai-IN",
32
+ DoiIn: "doi-IN",
14
33
  };
@@ -7,4 +7,17 @@ export interface SpeechToTextTranslateResponse {
7
7
  language_code?: SarvamAI.SpeechToTextTranslateLanguage;
8
8
  /** Diarized transcript of the provided speech */
9
9
  diarized_transcript?: SarvamAI.DiarizedTranscript;
10
+ /**
11
+ * Float value (0.0 to 1.0) indicating the probability of the detected language being correct. Higher values indicate higher confidence.
12
+ *
13
+ * **When it returns a value:**
14
+ * - When `language_code` is not provided in the request
15
+ * - When `language_code` is set to `unknown`
16
+ *
17
+ * **When it returns null:**
18
+ * - When a specific `language_code` is provided (language detection is skipped)
19
+ *
20
+ * The parameter is always present in the response.
21
+ */
22
+ language_probability?: number;
10
23
  }
@@ -6,5 +6,18 @@ export interface SpeechToTextTranslateTranscriptionData {
6
6
  transcript: string;
7
7
  /** BCP-47 code of detected source language (null when language detection is in progress) */
8
8
  language_code?: string;
9
+ /**
10
+ * Float value (0.0 to 1.0) indicating the probability of the detected language being correct. Higher values indicate higher confidence.
11
+ *
12
+ * **When it returns a value:**
13
+ * - When `language_code` is not provided in the request
14
+ * - When `language_code` is set to `unknown`
15
+ *
16
+ * **When it returns null:**
17
+ * - When a specific `language_code` is provided (language detection is skipped)
18
+ *
19
+ * The parameter is always present in the response.
20
+ */
21
+ language_probability?: number;
9
22
  metrics: SarvamAI.TranscriptionMetrics;
10
23
  }
@@ -1,5 +1,5 @@
1
1
  export declare const TextToSpeechModel: {
2
- readonly BulbulV3: "bulbul:v3";
3
2
  readonly BulbulV2: "bulbul:v2";
3
+ readonly BulbulV3: "bulbul:v3";
4
4
  };
5
5
  export type TextToSpeechModel = (typeof TextToSpeechModel)[keyof typeof TextToSpeechModel];
@@ -1,5 +1,5 @@
1
1
  // This file was auto-generated by Fern from our API Definition.
2
2
  export const TextToSpeechModel = {
3
- BulbulV3: "bulbul:v3",
4
3
  BulbulV2: "bulbul:v2",
4
+ BulbulV3: "bulbul:v3",
5
5
  };
@@ -1 +1 @@
1
- export declare const SDK_VERSION = "1.1.3";
1
+ export declare const SDK_VERSION = "1.1.4";
@@ -1 +1 @@
1
- export const SDK_VERSION = "1.1.3";
1
+ export const SDK_VERSION = "1.1.4";
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "sarvamai",
3
- "version": "1.1.3",
3
+ "version": "1.1.4",
4
4
  "private": false,
5
5
  "repository": "github:sarvamai/sarvam-javascript-sdk",
6
6
  "type": "commonjs",