sarvamai 1.1.5-alpha.2 → 1.1.5
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/dist/cjs/BaseClient.js +2 -2
- package/dist/cjs/api/resources/speechToTextStreaming/types/SpeechToTextStreamingLanguageCode.d.ts +2 -2
- package/dist/cjs/api/resources/speechToTextStreaming/types/SpeechToTextStreamingLanguageCode.js +2 -2
- package/dist/cjs/api/resources/speechToTextStreaming/types/SpeechToTextStreamingModel.d.ts +3 -3
- package/dist/cjs/api/resources/speechToTextStreaming/types/SpeechToTextStreamingModel.js +3 -3
- package/dist/cjs/api/resources/speechToTextTranslateStreaming/types/SpeechToTextTranslateStreamingMode.d.ts +22 -0
- package/dist/cjs/api/resources/speechToTextTranslateStreaming/types/SpeechToTextTranslateStreamingMode.js +25 -0
- package/dist/cjs/api/resources/speechToTextTranslateStreaming/types/SpeechToTextTranslateStreamingModel.d.ts +13 -0
- package/dist/cjs/api/resources/speechToTextTranslateStreaming/types/SpeechToTextTranslateStreamingModel.js +16 -0
- package/dist/cjs/api/resources/speechToTextTranslateStreaming/types/index.d.ts +2 -0
- package/dist/cjs/api/resources/speechToTextTranslateStreaming/types/index.js +2 -0
- package/dist/cjs/api/resources/textToSpeechStreaming/types/TextToSpeechStreamingModel.d.ts +2 -2
- package/dist/cjs/api/resources/textToSpeechStreaming/types/TextToSpeechStreamingModel.js +2 -2
- package/dist/cjs/api/types/ConfigureConnection.d.ts +12 -12
- package/dist/cjs/api/types/ConfigureConnection.js +3 -3
- package/dist/cjs/version.d.ts +1 -1
- package/dist/cjs/version.js +1 -1
- package/dist/esm/BaseClient.mjs +2 -2
- package/dist/esm/api/resources/speechToTextStreaming/types/SpeechToTextStreamingLanguageCode.d.mts +2 -2
- package/dist/esm/api/resources/speechToTextStreaming/types/SpeechToTextStreamingLanguageCode.mjs +2 -2
- package/dist/esm/api/resources/speechToTextStreaming/types/SpeechToTextStreamingModel.d.mts +3 -3
- package/dist/esm/api/resources/speechToTextStreaming/types/SpeechToTextStreamingModel.mjs +3 -3
- package/dist/esm/api/resources/speechToTextTranslateStreaming/types/SpeechToTextTranslateStreamingMode.d.mts +22 -0
- package/dist/esm/api/resources/speechToTextTranslateStreaming/types/SpeechToTextTranslateStreamingMode.mjs +22 -0
- package/dist/esm/api/resources/speechToTextTranslateStreaming/types/SpeechToTextTranslateStreamingModel.d.mts +13 -0
- package/dist/esm/api/resources/speechToTextTranslateStreaming/types/SpeechToTextTranslateStreamingModel.mjs +13 -0
- package/dist/esm/api/resources/speechToTextTranslateStreaming/types/index.d.mts +2 -0
- package/dist/esm/api/resources/speechToTextTranslateStreaming/types/index.mjs +2 -0
- package/dist/esm/api/resources/textToSpeechStreaming/types/TextToSpeechStreamingModel.d.mts +2 -2
- package/dist/esm/api/resources/textToSpeechStreaming/types/TextToSpeechStreamingModel.mjs +2 -2
- package/dist/esm/api/types/ConfigureConnection.d.mts +12 -12
- package/dist/esm/api/types/ConfigureConnection.mjs +3 -3
- package/dist/esm/version.d.mts +1 -1
- package/dist/esm/version.mjs +1 -1
- package/package.json +1 -1
package/dist/cjs/BaseClient.js
CHANGED
|
@@ -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": "1.1.5
|
|
47
|
-
"User-Agent": "sarvamai/1.1.5
|
|
46
|
+
"X-Fern-SDK-Version": "1.1.5",
|
|
47
|
+
"User-Agent": "sarvamai/1.1.5",
|
|
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);
|
package/dist/cjs/api/resources/speechToTextStreaming/types/SpeechToTextStreamingLanguageCode.d.ts
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
/**
|
|
2
2
|
* Specifies the language of the input audio in BCP-47 format.
|
|
3
3
|
*
|
|
4
|
-
* **Available Options (saarika:v2.5):**
|
|
4
|
+
* **Available Options (saarika:v2.5, legacy):**
|
|
5
5
|
* - `unknown` (default): Use when the language is not known; the API will auto-detect.
|
|
6
6
|
* - `hi-IN`: Hindi
|
|
7
7
|
* - `bn-IN`: Bengali
|
|
@@ -15,7 +15,7 @@
|
|
|
15
15
|
* - `te-IN`: Telugu
|
|
16
16
|
* - `en-IN`: English
|
|
17
17
|
*
|
|
18
|
-
* **Additional Options (saaras:v3
|
|
18
|
+
* **Additional Options (saaras:v3, recommended):**
|
|
19
19
|
* - `as-IN`: Assamese
|
|
20
20
|
* - `ur-IN`: Urdu
|
|
21
21
|
* - `ne-IN`: Nepali
|
package/dist/cjs/api/resources/speechToTextStreaming/types/SpeechToTextStreamingLanguageCode.js
CHANGED
|
@@ -5,7 +5,7 @@ exports.SpeechToTextStreamingLanguageCode = void 0;
|
|
|
5
5
|
/**
|
|
6
6
|
* Specifies the language of the input audio in BCP-47 format.
|
|
7
7
|
*
|
|
8
|
-
* **Available Options (saarika:v2.5):**
|
|
8
|
+
* **Available Options (saarika:v2.5, legacy):**
|
|
9
9
|
* - `unknown` (default): Use when the language is not known; the API will auto-detect.
|
|
10
10
|
* - `hi-IN`: Hindi
|
|
11
11
|
* - `bn-IN`: Bengali
|
|
@@ -19,7 +19,7 @@ exports.SpeechToTextStreamingLanguageCode = void 0;
|
|
|
19
19
|
* - `te-IN`: Telugu
|
|
20
20
|
* - `en-IN`: English
|
|
21
21
|
*
|
|
22
|
-
* **Additional Options (saaras:v3
|
|
22
|
+
* **Additional Options (saaras:v3, recommended):**
|
|
23
23
|
* - `as-IN`: Assamese
|
|
24
24
|
* - `ur-IN`: Urdu
|
|
25
25
|
* - `ne-IN`: Nepali
|
|
@@ -1,12 +1,12 @@
|
|
|
1
1
|
/**
|
|
2
2
|
* Specifies the model to use for speech-to-text conversion.
|
|
3
3
|
*
|
|
4
|
-
* - **
|
|
4
|
+
* - **saaras:v3** (default, recommended): State-of-the-art model with flexible output formats. Supports multiple modes via the `mode` parameter: transcribe, translate, verbatim, translit, codemix.
|
|
5
5
|
*
|
|
6
|
-
* - **
|
|
6
|
+
* - **saarika:v2.5** (legacy): Transcribes audio in the spoken language. Kept for backward compatibility.
|
|
7
7
|
*/
|
|
8
8
|
export declare const SpeechToTextStreamingModel: {
|
|
9
|
-
readonly SaarikaV25: "saarika:v2.5";
|
|
10
9
|
readonly SaarasV3: "saaras:v3";
|
|
10
|
+
readonly SaarikaV25: "saarika:v2.5";
|
|
11
11
|
};
|
|
12
12
|
export type SpeechToTextStreamingModel = (typeof SpeechToTextStreamingModel)[keyof typeof SpeechToTextStreamingModel];
|
|
@@ -5,11 +5,11 @@ exports.SpeechToTextStreamingModel = void 0;
|
|
|
5
5
|
/**
|
|
6
6
|
* Specifies the model to use for speech-to-text conversion.
|
|
7
7
|
*
|
|
8
|
-
* - **
|
|
8
|
+
* - **saaras:v3** (default, recommended): State-of-the-art model with flexible output formats. Supports multiple modes via the `mode` parameter: transcribe, translate, verbatim, translit, codemix.
|
|
9
9
|
*
|
|
10
|
-
* - **
|
|
10
|
+
* - **saarika:v2.5** (legacy): Transcribes audio in the spoken language. Kept for backward compatibility.
|
|
11
11
|
*/
|
|
12
12
|
exports.SpeechToTextStreamingModel = {
|
|
13
|
-
SaarikaV25: "saarika:v2.5",
|
|
14
13
|
SaarasV3: "saaras:v3",
|
|
14
|
+
SaarikaV25: "saarika:v2.5",
|
|
15
15
|
};
|
|
@@ -0,0 +1,22 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* Mode of operation. **Only applicable when using saaras:v3 model.**
|
|
3
|
+
*
|
|
4
|
+
* - **translate** (default): Translates speech from any supported Indic language to English.
|
|
5
|
+
* - Example: Hindi audio → English text output
|
|
6
|
+
*
|
|
7
|
+
* - **transcribe**: Standard transcription in the original language.
|
|
8
|
+
*
|
|
9
|
+
* - **verbatim**: Exact word-for-word transcription without normalization.
|
|
10
|
+
*
|
|
11
|
+
* - **translit**: Romanization - Transliterates speech to Latin/Roman script only.
|
|
12
|
+
*
|
|
13
|
+
* - **codemix**: Code-mixed text with English words in English and Indic words in native script.
|
|
14
|
+
*/
|
|
15
|
+
export declare const SpeechToTextTranslateStreamingMode: {
|
|
16
|
+
readonly Translate: "translate";
|
|
17
|
+
readonly Transcribe: "transcribe";
|
|
18
|
+
readonly Verbatim: "verbatim";
|
|
19
|
+
readonly Translit: "translit";
|
|
20
|
+
readonly Codemix: "codemix";
|
|
21
|
+
};
|
|
22
|
+
export type SpeechToTextTranslateStreamingMode = (typeof SpeechToTextTranslateStreamingMode)[keyof typeof SpeechToTextTranslateStreamingMode];
|
|
@@ -0,0 +1,25 @@
|
|
|
1
|
+
"use strict";
|
|
2
|
+
// This file was auto-generated by Fern from our API Definition.
|
|
3
|
+
Object.defineProperty(exports, "__esModule", { value: true });
|
|
4
|
+
exports.SpeechToTextTranslateStreamingMode = void 0;
|
|
5
|
+
/**
|
|
6
|
+
* Mode of operation. **Only applicable when using saaras:v3 model.**
|
|
7
|
+
*
|
|
8
|
+
* - **translate** (default): Translates speech from any supported Indic language to English.
|
|
9
|
+
* - Example: Hindi audio → English text output
|
|
10
|
+
*
|
|
11
|
+
* - **transcribe**: Standard transcription in the original language.
|
|
12
|
+
*
|
|
13
|
+
* - **verbatim**: Exact word-for-word transcription without normalization.
|
|
14
|
+
*
|
|
15
|
+
* - **translit**: Romanization - Transliterates speech to Latin/Roman script only.
|
|
16
|
+
*
|
|
17
|
+
* - **codemix**: Code-mixed text with English words in English and Indic words in native script.
|
|
18
|
+
*/
|
|
19
|
+
exports.SpeechToTextTranslateStreamingMode = {
|
|
20
|
+
Translate: "translate",
|
|
21
|
+
Transcribe: "transcribe",
|
|
22
|
+
Verbatim: "verbatim",
|
|
23
|
+
Translit: "translit",
|
|
24
|
+
Codemix: "codemix",
|
|
25
|
+
};
|
|
@@ -0,0 +1,13 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* Model to be used for speech to text translation.
|
|
3
|
+
*
|
|
4
|
+
* - **saaras:v3** (default, recommended): State-of-the-art translation model that translates audio from any spoken Indic language to English with flexible output formats via the `mode` parameter.
|
|
5
|
+
*
|
|
6
|
+
* - **saaras:v2.5** (legacy): Translation model that translates audio from any spoken Indic language to English. Kept for backward compatibility.
|
|
7
|
+
* - Example: Hindi audio → English text output
|
|
8
|
+
*/
|
|
9
|
+
export declare const SpeechToTextTranslateStreamingModel: {
|
|
10
|
+
readonly SaarasV3: "saaras:v3";
|
|
11
|
+
readonly SaarasV25: "saaras:v2.5";
|
|
12
|
+
};
|
|
13
|
+
export type SpeechToTextTranslateStreamingModel = (typeof SpeechToTextTranslateStreamingModel)[keyof typeof SpeechToTextTranslateStreamingModel];
|
|
@@ -0,0 +1,16 @@
|
|
|
1
|
+
"use strict";
|
|
2
|
+
// This file was auto-generated by Fern from our API Definition.
|
|
3
|
+
Object.defineProperty(exports, "__esModule", { value: true });
|
|
4
|
+
exports.SpeechToTextTranslateStreamingModel = void 0;
|
|
5
|
+
/**
|
|
6
|
+
* Model to be used for speech to text translation.
|
|
7
|
+
*
|
|
8
|
+
* - **saaras:v3** (default, recommended): State-of-the-art translation model that translates audio from any spoken Indic language to English with flexible output formats via the `mode` parameter.
|
|
9
|
+
*
|
|
10
|
+
* - **saaras:v2.5** (legacy): Translation model that translates audio from any spoken Indic language to English. Kept for backward compatibility.
|
|
11
|
+
* - Example: Hindi audio → English text output
|
|
12
|
+
*/
|
|
13
|
+
exports.SpeechToTextTranslateStreamingModel = {
|
|
14
|
+
SaarasV3: "saaras:v3",
|
|
15
|
+
SaarasV25: "saaras:v2.5",
|
|
16
|
+
};
|
|
@@ -1,4 +1,6 @@
|
|
|
1
1
|
export * from "./SpeechToTextTranslateStreamingFlushSignal.js";
|
|
2
2
|
export * from "./SpeechToTextTranslateStreamingHighVadSensitivity.js";
|
|
3
3
|
export * from "./SpeechToTextTranslateStreamingInputAudioCodec.js";
|
|
4
|
+
export * from "./SpeechToTextTranslateStreamingMode.js";
|
|
5
|
+
export * from "./SpeechToTextTranslateStreamingModel.js";
|
|
4
6
|
export * from "./SpeechToTextTranslateStreamingVadSignals.js";
|
|
@@ -17,4 +17,6 @@ Object.defineProperty(exports, "__esModule", { value: true });
|
|
|
17
17
|
__exportStar(require("./SpeechToTextTranslateStreamingFlushSignal.js"), exports);
|
|
18
18
|
__exportStar(require("./SpeechToTextTranslateStreamingHighVadSensitivity.js"), exports);
|
|
19
19
|
__exportStar(require("./SpeechToTextTranslateStreamingInputAudioCodec.js"), exports);
|
|
20
|
+
__exportStar(require("./SpeechToTextTranslateStreamingMode.js"), exports);
|
|
21
|
+
__exportStar(require("./SpeechToTextTranslateStreamingModel.js"), exports);
|
|
20
22
|
__exportStar(require("./SpeechToTextTranslateStreamingVadSignals.js"), exports);
|
|
@@ -1,10 +1,10 @@
|
|
|
1
1
|
/**
|
|
2
2
|
* Text to speech model to use.
|
|
3
3
|
* - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
|
|
4
|
-
* - **bulbul:v3
|
|
4
|
+
* - **bulbul:v3**: Advanced model with temperature control (no pitch/loudness)
|
|
5
5
|
*/
|
|
6
6
|
export declare const TextToSpeechStreamingModel: {
|
|
7
7
|
readonly BulbulV2: "bulbul:v2";
|
|
8
|
-
readonly
|
|
8
|
+
readonly BulbulV3: "bulbul:v3";
|
|
9
9
|
};
|
|
10
10
|
export type TextToSpeechStreamingModel = (typeof TextToSpeechStreamingModel)[keyof typeof TextToSpeechStreamingModel];
|
|
@@ -5,9 +5,9 @@ exports.TextToSpeechStreamingModel = void 0;
|
|
|
5
5
|
/**
|
|
6
6
|
* Text to speech model to use.
|
|
7
7
|
* - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
|
|
8
|
-
* - **bulbul:v3
|
|
8
|
+
* - **bulbul:v3**: Advanced model with temperature control (no pitch/loudness)
|
|
9
9
|
*/
|
|
10
10
|
exports.TextToSpeechStreamingModel = {
|
|
11
11
|
BulbulV2: "bulbul:v2",
|
|
12
|
-
|
|
12
|
+
BulbulV3: "bulbul:v3",
|
|
13
13
|
};
|
|
@@ -6,7 +6,7 @@
|
|
|
6
6
|
*
|
|
7
7
|
* **Model-Specific Notes:**
|
|
8
8
|
* - **bulbul:v2:** Supports pitch, loudness, pace (0.3-3.0). Default sample rate: 22050 Hz.
|
|
9
|
-
* - **bulbul:v3
|
|
9
|
+
* - **bulbul:v3:** 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";
|
|
@@ -17,7 +17,7 @@ export declare namespace ConfigureConnection {
|
|
|
17
17
|
/**
|
|
18
18
|
* Specifies the model to use for text-to-speech conversion.
|
|
19
19
|
* - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
|
|
20
|
-
* - **bulbul:v3
|
|
20
|
+
* - **bulbul:v3**: Advanced model with temperature control (no pitch/loudness)
|
|
21
21
|
*/
|
|
22
22
|
model?: Data.Model;
|
|
23
23
|
/** The language of the text in BCP-47 format */
|
|
@@ -27,7 +27,7 @@ export declare namespace ConfigureConnection {
|
|
|
27
27
|
*
|
|
28
28
|
* **Model Compatibility:**
|
|
29
29
|
* - **bulbul:v2:** anushka (default), abhilash, manisha, vidya, arya, karun, hitesh
|
|
30
|
-
* - **bulbul:v3
|
|
30
|
+
* - **bulbul:v3:** 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
|
|
40
|
+
* **Note:** NOT supported for bulbul:v3. Will be ignored if provided.
|
|
41
41
|
*/
|
|
42
42
|
pitch?: number;
|
|
43
43
|
/**
|
|
@@ -46,7 +46,7 @@ export declare namespace ConfigureConnection {
|
|
|
46
46
|
*
|
|
47
47
|
* **Model-specific ranges:**
|
|
48
48
|
* - **bulbul:v2:** 0.3 to 3.0
|
|
49
|
-
* - **bulbul:v3
|
|
49
|
+
* - **bulbul:v3:** 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
|
|
57
|
+
* **Note:** NOT supported for bulbul:v3. Will be ignored if provided.
|
|
58
58
|
*/
|
|
59
59
|
loudness?: number;
|
|
60
60
|
/**
|
|
@@ -62,7 +62,7 @@ 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
|
|
65
|
+
* **Note:** Only supported for bulbul:v3. Will be ignored for bulbul:v2.
|
|
66
66
|
*/
|
|
67
67
|
temperature?: number;
|
|
68
68
|
/**
|
|
@@ -71,7 +71,7 @@ export declare namespace ConfigureConnection {
|
|
|
71
71
|
*
|
|
72
72
|
* **Model-specific defaults:**
|
|
73
73
|
* - **bulbul:v2:** 22050 Hz
|
|
74
|
-
* - **bulbul:v3
|
|
74
|
+
* - **bulbul:v3:** 24000 Hz
|
|
75
75
|
*/
|
|
76
76
|
speech_sample_rate?: number;
|
|
77
77
|
/**
|
|
@@ -81,7 +81,7 @@ export declare namespace ConfigureConnection {
|
|
|
81
81
|
*
|
|
82
82
|
* **Model-specific defaults:**
|
|
83
83
|
* - **bulbul:v2:** false (optional)
|
|
84
|
-
* - **bulbul:v3
|
|
84
|
+
* - **bulbul:v3:** Always enabled (cannot be disabled)
|
|
85
85
|
*/
|
|
86
86
|
enable_preprocessing?: boolean;
|
|
87
87
|
/** Audio codec (currently supports MP3 only, optimized for real-time playback) */
|
|
@@ -97,11 +97,11 @@ export declare namespace ConfigureConnection {
|
|
|
97
97
|
/**
|
|
98
98
|
* Specifies the model to use for text-to-speech conversion.
|
|
99
99
|
* - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
|
|
100
|
-
* - **bulbul:v3
|
|
100
|
+
* - **bulbul:v3**: Advanced model with temperature control (no pitch/loudness)
|
|
101
101
|
*/
|
|
102
102
|
const Model: {
|
|
103
103
|
readonly BulbulV2: "bulbul:v2";
|
|
104
|
-
readonly
|
|
104
|
+
readonly BulbulV3: "bulbul:v3";
|
|
105
105
|
};
|
|
106
106
|
type Model = (typeof Model)[keyof typeof Model];
|
|
107
107
|
/** The language of the text in BCP-47 format */
|
|
@@ -124,7 +124,7 @@ export declare namespace ConfigureConnection {
|
|
|
124
124
|
*
|
|
125
125
|
* **Model Compatibility:**
|
|
126
126
|
* - **bulbul:v2:** anushka (default), abhilash, manisha, vidya, arya, karun, hitesh
|
|
127
|
-
* - **bulbul:v3
|
|
127
|
+
* - **bulbul:v3:** 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
|
|
128
128
|
*
|
|
129
129
|
* **Note:** Speaker selection must match the chosen model version.
|
|
130
130
|
*/
|
|
@@ -9,11 +9,11 @@ var ConfigureConnection;
|
|
|
9
9
|
/**
|
|
10
10
|
* Specifies the model to use for text-to-speech conversion.
|
|
11
11
|
* - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
|
|
12
|
-
* - **bulbul:v3
|
|
12
|
+
* - **bulbul:v3**: Advanced model with temperature control (no pitch/loudness)
|
|
13
13
|
*/
|
|
14
14
|
Data.Model = {
|
|
15
15
|
BulbulV2: "bulbul:v2",
|
|
16
|
-
|
|
16
|
+
BulbulV3: "bulbul:v3",
|
|
17
17
|
};
|
|
18
18
|
/** The language of the text in BCP-47 format */
|
|
19
19
|
Data.TargetLanguageCode = {
|
|
@@ -34,7 +34,7 @@ var ConfigureConnection;
|
|
|
34
34
|
*
|
|
35
35
|
* **Model Compatibility:**
|
|
36
36
|
* - **bulbul:v2:** anushka (default), abhilash, manisha, vidya, arya, karun, hitesh
|
|
37
|
-
* - **bulbul:v3
|
|
37
|
+
* - **bulbul:v3:** 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
|
*/
|
package/dist/cjs/version.d.ts
CHANGED
|
@@ -1 +1 @@
|
|
|
1
|
-
export declare const SDK_VERSION = "1.1.5
|
|
1
|
+
export declare const SDK_VERSION = "1.1.5";
|
package/dist/cjs/version.js
CHANGED
package/dist/esm/BaseClient.mjs
CHANGED
|
@@ -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": "1.1.5
|
|
10
|
-
"User-Agent": "sarvamai/1.1.5
|
|
9
|
+
"X-Fern-SDK-Version": "1.1.5",
|
|
10
|
+
"User-Agent": "sarvamai/1.1.5",
|
|
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);
|
package/dist/esm/api/resources/speechToTextStreaming/types/SpeechToTextStreamingLanguageCode.d.mts
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
/**
|
|
2
2
|
* Specifies the language of the input audio in BCP-47 format.
|
|
3
3
|
*
|
|
4
|
-
* **Available Options (saarika:v2.5):**
|
|
4
|
+
* **Available Options (saarika:v2.5, legacy):**
|
|
5
5
|
* - `unknown` (default): Use when the language is not known; the API will auto-detect.
|
|
6
6
|
* - `hi-IN`: Hindi
|
|
7
7
|
* - `bn-IN`: Bengali
|
|
@@ -15,7 +15,7 @@
|
|
|
15
15
|
* - `te-IN`: Telugu
|
|
16
16
|
* - `en-IN`: English
|
|
17
17
|
*
|
|
18
|
-
* **Additional Options (saaras:v3
|
|
18
|
+
* **Additional Options (saaras:v3, recommended):**
|
|
19
19
|
* - `as-IN`: Assamese
|
|
20
20
|
* - `ur-IN`: Urdu
|
|
21
21
|
* - `ne-IN`: Nepali
|
package/dist/esm/api/resources/speechToTextStreaming/types/SpeechToTextStreamingLanguageCode.mjs
CHANGED
|
@@ -2,7 +2,7 @@
|
|
|
2
2
|
/**
|
|
3
3
|
* Specifies the language of the input audio in BCP-47 format.
|
|
4
4
|
*
|
|
5
|
-
* **Available Options (saarika:v2.5):**
|
|
5
|
+
* **Available Options (saarika:v2.5, legacy):**
|
|
6
6
|
* - `unknown` (default): Use when the language is not known; the API will auto-detect.
|
|
7
7
|
* - `hi-IN`: Hindi
|
|
8
8
|
* - `bn-IN`: Bengali
|
|
@@ -16,7 +16,7 @@
|
|
|
16
16
|
* - `te-IN`: Telugu
|
|
17
17
|
* - `en-IN`: English
|
|
18
18
|
*
|
|
19
|
-
* **Additional Options (saaras:v3
|
|
19
|
+
* **Additional Options (saaras:v3, recommended):**
|
|
20
20
|
* - `as-IN`: Assamese
|
|
21
21
|
* - `ur-IN`: Urdu
|
|
22
22
|
* - `ne-IN`: Nepali
|
|
@@ -1,12 +1,12 @@
|
|
|
1
1
|
/**
|
|
2
2
|
* Specifies the model to use for speech-to-text conversion.
|
|
3
3
|
*
|
|
4
|
-
* - **
|
|
4
|
+
* - **saaras:v3** (default, recommended): State-of-the-art model with flexible output formats. Supports multiple modes via the `mode` parameter: transcribe, translate, verbatim, translit, codemix.
|
|
5
5
|
*
|
|
6
|
-
* - **
|
|
6
|
+
* - **saarika:v2.5** (legacy): Transcribes audio in the spoken language. Kept for backward compatibility.
|
|
7
7
|
*/
|
|
8
8
|
export declare const SpeechToTextStreamingModel: {
|
|
9
|
-
readonly SaarikaV25: "saarika:v2.5";
|
|
10
9
|
readonly SaarasV3: "saaras:v3";
|
|
10
|
+
readonly SaarikaV25: "saarika:v2.5";
|
|
11
11
|
};
|
|
12
12
|
export type SpeechToTextStreamingModel = (typeof SpeechToTextStreamingModel)[keyof typeof SpeechToTextStreamingModel];
|
|
@@ -2,11 +2,11 @@
|
|
|
2
2
|
/**
|
|
3
3
|
* Specifies the model to use for speech-to-text conversion.
|
|
4
4
|
*
|
|
5
|
-
* - **
|
|
5
|
+
* - **saaras:v3** (default, recommended): State-of-the-art model with flexible output formats. Supports multiple modes via the `mode` parameter: transcribe, translate, verbatim, translit, codemix.
|
|
6
6
|
*
|
|
7
|
-
* - **
|
|
7
|
+
* - **saarika:v2.5** (legacy): Transcribes audio in the spoken language. Kept for backward compatibility.
|
|
8
8
|
*/
|
|
9
9
|
export const SpeechToTextStreamingModel = {
|
|
10
|
-
SaarikaV25: "saarika:v2.5",
|
|
11
10
|
SaarasV3: "saaras:v3",
|
|
11
|
+
SaarikaV25: "saarika:v2.5",
|
|
12
12
|
};
|
|
@@ -0,0 +1,22 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* Mode of operation. **Only applicable when using saaras:v3 model.**
|
|
3
|
+
*
|
|
4
|
+
* - **translate** (default): Translates speech from any supported Indic language to English.
|
|
5
|
+
* - Example: Hindi audio → English text output
|
|
6
|
+
*
|
|
7
|
+
* - **transcribe**: Standard transcription in the original language.
|
|
8
|
+
*
|
|
9
|
+
* - **verbatim**: Exact word-for-word transcription without normalization.
|
|
10
|
+
*
|
|
11
|
+
* - **translit**: Romanization - Transliterates speech to Latin/Roman script only.
|
|
12
|
+
*
|
|
13
|
+
* - **codemix**: Code-mixed text with English words in English and Indic words in native script.
|
|
14
|
+
*/
|
|
15
|
+
export declare const SpeechToTextTranslateStreamingMode: {
|
|
16
|
+
readonly Translate: "translate";
|
|
17
|
+
readonly Transcribe: "transcribe";
|
|
18
|
+
readonly Verbatim: "verbatim";
|
|
19
|
+
readonly Translit: "translit";
|
|
20
|
+
readonly Codemix: "codemix";
|
|
21
|
+
};
|
|
22
|
+
export type SpeechToTextTranslateStreamingMode = (typeof SpeechToTextTranslateStreamingMode)[keyof typeof SpeechToTextTranslateStreamingMode];
|
|
@@ -0,0 +1,22 @@
|
|
|
1
|
+
// This file was auto-generated by Fern from our API Definition.
|
|
2
|
+
/**
|
|
3
|
+
* Mode of operation. **Only applicable when using saaras:v3 model.**
|
|
4
|
+
*
|
|
5
|
+
* - **translate** (default): Translates speech from any supported Indic language to English.
|
|
6
|
+
* - Example: Hindi audio → English text output
|
|
7
|
+
*
|
|
8
|
+
* - **transcribe**: Standard transcription in the original language.
|
|
9
|
+
*
|
|
10
|
+
* - **verbatim**: Exact word-for-word transcription without normalization.
|
|
11
|
+
*
|
|
12
|
+
* - **translit**: Romanization - Transliterates speech to Latin/Roman script only.
|
|
13
|
+
*
|
|
14
|
+
* - **codemix**: Code-mixed text with English words in English and Indic words in native script.
|
|
15
|
+
*/
|
|
16
|
+
export const SpeechToTextTranslateStreamingMode = {
|
|
17
|
+
Translate: "translate",
|
|
18
|
+
Transcribe: "transcribe",
|
|
19
|
+
Verbatim: "verbatim",
|
|
20
|
+
Translit: "translit",
|
|
21
|
+
Codemix: "codemix",
|
|
22
|
+
};
|
|
@@ -0,0 +1,13 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* Model to be used for speech to text translation.
|
|
3
|
+
*
|
|
4
|
+
* - **saaras:v3** (default, recommended): State-of-the-art translation model that translates audio from any spoken Indic language to English with flexible output formats via the `mode` parameter.
|
|
5
|
+
*
|
|
6
|
+
* - **saaras:v2.5** (legacy): Translation model that translates audio from any spoken Indic language to English. Kept for backward compatibility.
|
|
7
|
+
* - Example: Hindi audio → English text output
|
|
8
|
+
*/
|
|
9
|
+
export declare const SpeechToTextTranslateStreamingModel: {
|
|
10
|
+
readonly SaarasV3: "saaras:v3";
|
|
11
|
+
readonly SaarasV25: "saaras:v2.5";
|
|
12
|
+
};
|
|
13
|
+
export type SpeechToTextTranslateStreamingModel = (typeof SpeechToTextTranslateStreamingModel)[keyof typeof SpeechToTextTranslateStreamingModel];
|
|
@@ -0,0 +1,13 @@
|
|
|
1
|
+
// This file was auto-generated by Fern from our API Definition.
|
|
2
|
+
/**
|
|
3
|
+
* Model to be used for speech to text translation.
|
|
4
|
+
*
|
|
5
|
+
* - **saaras:v3** (default, recommended): State-of-the-art translation model that translates audio from any spoken Indic language to English with flexible output formats via the `mode` parameter.
|
|
6
|
+
*
|
|
7
|
+
* - **saaras:v2.5** (legacy): Translation model that translates audio from any spoken Indic language to English. Kept for backward compatibility.
|
|
8
|
+
* - Example: Hindi audio → English text output
|
|
9
|
+
*/
|
|
10
|
+
export const SpeechToTextTranslateStreamingModel = {
|
|
11
|
+
SaarasV3: "saaras:v3",
|
|
12
|
+
SaarasV25: "saaras:v2.5",
|
|
13
|
+
};
|
|
@@ -1,4 +1,6 @@
|
|
|
1
1
|
export * from "./SpeechToTextTranslateStreamingFlushSignal.mjs";
|
|
2
2
|
export * from "./SpeechToTextTranslateStreamingHighVadSensitivity.mjs";
|
|
3
3
|
export * from "./SpeechToTextTranslateStreamingInputAudioCodec.mjs";
|
|
4
|
+
export * from "./SpeechToTextTranslateStreamingMode.mjs";
|
|
5
|
+
export * from "./SpeechToTextTranslateStreamingModel.mjs";
|
|
4
6
|
export * from "./SpeechToTextTranslateStreamingVadSignals.mjs";
|
|
@@ -1,4 +1,6 @@
|
|
|
1
1
|
export * from "./SpeechToTextTranslateStreamingFlushSignal.mjs";
|
|
2
2
|
export * from "./SpeechToTextTranslateStreamingHighVadSensitivity.mjs";
|
|
3
3
|
export * from "./SpeechToTextTranslateStreamingInputAudioCodec.mjs";
|
|
4
|
+
export * from "./SpeechToTextTranslateStreamingMode.mjs";
|
|
5
|
+
export * from "./SpeechToTextTranslateStreamingModel.mjs";
|
|
4
6
|
export * from "./SpeechToTextTranslateStreamingVadSignals.mjs";
|
|
@@ -1,10 +1,10 @@
|
|
|
1
1
|
/**
|
|
2
2
|
* Text to speech model to use.
|
|
3
3
|
* - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
|
|
4
|
-
* - **bulbul:v3
|
|
4
|
+
* - **bulbul:v3**: Advanced model with temperature control (no pitch/loudness)
|
|
5
5
|
*/
|
|
6
6
|
export declare const TextToSpeechStreamingModel: {
|
|
7
7
|
readonly BulbulV2: "bulbul:v2";
|
|
8
|
-
readonly
|
|
8
|
+
readonly BulbulV3: "bulbul:v3";
|
|
9
9
|
};
|
|
10
10
|
export type TextToSpeechStreamingModel = (typeof TextToSpeechStreamingModel)[keyof typeof TextToSpeechStreamingModel];
|
|
@@ -2,9 +2,9 @@
|
|
|
2
2
|
/**
|
|
3
3
|
* Text to speech model to use.
|
|
4
4
|
* - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
|
|
5
|
-
* - **bulbul:v3
|
|
5
|
+
* - **bulbul:v3**: Advanced model with temperature control (no pitch/loudness)
|
|
6
6
|
*/
|
|
7
7
|
export const TextToSpeechStreamingModel = {
|
|
8
8
|
BulbulV2: "bulbul:v2",
|
|
9
|
-
|
|
9
|
+
BulbulV3: "bulbul:v3",
|
|
10
10
|
};
|
|
@@ -6,7 +6,7 @@
|
|
|
6
6
|
*
|
|
7
7
|
* **Model-Specific Notes:**
|
|
8
8
|
* - **bulbul:v2:** Supports pitch, loudness, pace (0.3-3.0). Default sample rate: 22050 Hz.
|
|
9
|
-
* - **bulbul:v3
|
|
9
|
+
* - **bulbul:v3:** 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";
|
|
@@ -17,7 +17,7 @@ export declare namespace ConfigureConnection {
|
|
|
17
17
|
/**
|
|
18
18
|
* Specifies the model to use for text-to-speech conversion.
|
|
19
19
|
* - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
|
|
20
|
-
* - **bulbul:v3
|
|
20
|
+
* - **bulbul:v3**: Advanced model with temperature control (no pitch/loudness)
|
|
21
21
|
*/
|
|
22
22
|
model?: Data.Model;
|
|
23
23
|
/** The language of the text in BCP-47 format */
|
|
@@ -27,7 +27,7 @@ export declare namespace ConfigureConnection {
|
|
|
27
27
|
*
|
|
28
28
|
* **Model Compatibility:**
|
|
29
29
|
* - **bulbul:v2:** anushka (default), abhilash, manisha, vidya, arya, karun, hitesh
|
|
30
|
-
* - **bulbul:v3
|
|
30
|
+
* - **bulbul:v3:** 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
|
|
40
|
+
* **Note:** NOT supported for bulbul:v3. Will be ignored if provided.
|
|
41
41
|
*/
|
|
42
42
|
pitch?: number;
|
|
43
43
|
/**
|
|
@@ -46,7 +46,7 @@ export declare namespace ConfigureConnection {
|
|
|
46
46
|
*
|
|
47
47
|
* **Model-specific ranges:**
|
|
48
48
|
* - **bulbul:v2:** 0.3 to 3.0
|
|
49
|
-
* - **bulbul:v3
|
|
49
|
+
* - **bulbul:v3:** 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
|
|
57
|
+
* **Note:** NOT supported for bulbul:v3. Will be ignored if provided.
|
|
58
58
|
*/
|
|
59
59
|
loudness?: number;
|
|
60
60
|
/**
|
|
@@ -62,7 +62,7 @@ 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
|
|
65
|
+
* **Note:** Only supported for bulbul:v3. Will be ignored for bulbul:v2.
|
|
66
66
|
*/
|
|
67
67
|
temperature?: number;
|
|
68
68
|
/**
|
|
@@ -71,7 +71,7 @@ export declare namespace ConfigureConnection {
|
|
|
71
71
|
*
|
|
72
72
|
* **Model-specific defaults:**
|
|
73
73
|
* - **bulbul:v2:** 22050 Hz
|
|
74
|
-
* - **bulbul:v3
|
|
74
|
+
* - **bulbul:v3:** 24000 Hz
|
|
75
75
|
*/
|
|
76
76
|
speech_sample_rate?: number;
|
|
77
77
|
/**
|
|
@@ -81,7 +81,7 @@ export declare namespace ConfigureConnection {
|
|
|
81
81
|
*
|
|
82
82
|
* **Model-specific defaults:**
|
|
83
83
|
* - **bulbul:v2:** false (optional)
|
|
84
|
-
* - **bulbul:v3
|
|
84
|
+
* - **bulbul:v3:** Always enabled (cannot be disabled)
|
|
85
85
|
*/
|
|
86
86
|
enable_preprocessing?: boolean;
|
|
87
87
|
/** Audio codec (currently supports MP3 only, optimized for real-time playback) */
|
|
@@ -97,11 +97,11 @@ export declare namespace ConfigureConnection {
|
|
|
97
97
|
/**
|
|
98
98
|
* Specifies the model to use for text-to-speech conversion.
|
|
99
99
|
* - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
|
|
100
|
-
* - **bulbul:v3
|
|
100
|
+
* - **bulbul:v3**: Advanced model with temperature control (no pitch/loudness)
|
|
101
101
|
*/
|
|
102
102
|
const Model: {
|
|
103
103
|
readonly BulbulV2: "bulbul:v2";
|
|
104
|
-
readonly
|
|
104
|
+
readonly BulbulV3: "bulbul:v3";
|
|
105
105
|
};
|
|
106
106
|
type Model = (typeof Model)[keyof typeof Model];
|
|
107
107
|
/** The language of the text in BCP-47 format */
|
|
@@ -124,7 +124,7 @@ export declare namespace ConfigureConnection {
|
|
|
124
124
|
*
|
|
125
125
|
* **Model Compatibility:**
|
|
126
126
|
* - **bulbul:v2:** anushka (default), abhilash, manisha, vidya, arya, karun, hitesh
|
|
127
|
-
* - **bulbul:v3
|
|
127
|
+
* - **bulbul:v3:** 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
|
|
128
128
|
*
|
|
129
129
|
* **Note:** Speaker selection must match the chosen model version.
|
|
130
130
|
*/
|
|
@@ -6,11 +6,11 @@ export var ConfigureConnection;
|
|
|
6
6
|
/**
|
|
7
7
|
* Specifies the model to use for text-to-speech conversion.
|
|
8
8
|
* - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
|
|
9
|
-
* - **bulbul:v3
|
|
9
|
+
* - **bulbul:v3**: Advanced model with temperature control (no pitch/loudness)
|
|
10
10
|
*/
|
|
11
11
|
Data.Model = {
|
|
12
12
|
BulbulV2: "bulbul:v2",
|
|
13
|
-
|
|
13
|
+
BulbulV3: "bulbul:v3",
|
|
14
14
|
};
|
|
15
15
|
/** The language of the text in BCP-47 format */
|
|
16
16
|
Data.TargetLanguageCode = {
|
|
@@ -31,7 +31,7 @@ export var ConfigureConnection;
|
|
|
31
31
|
*
|
|
32
32
|
* **Model Compatibility:**
|
|
33
33
|
* - **bulbul:v2:** anushka (default), abhilash, manisha, vidya, arya, karun, hitesh
|
|
34
|
-
* - **bulbul:v3
|
|
34
|
+
* - **bulbul:v3:** 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
|
*/
|
package/dist/esm/version.d.mts
CHANGED
|
@@ -1 +1 @@
|
|
|
1
|
-
export declare const SDK_VERSION = "1.1.5
|
|
1
|
+
export declare const SDK_VERSION = "1.1.5";
|
package/dist/esm/version.mjs
CHANGED
|
@@ -1 +1 @@
|
|
|
1
|
-
export const SDK_VERSION = "1.1.5
|
|
1
|
+
export const SDK_VERSION = "1.1.5";
|