sarvamai 0.1.23a3__py3-none-any.whl → 0.1.23a5__py3-none-any.whl

sarvamai/speech_to_text/client.py

@@ -6,6 +6,7 @@ from .. import core
 from ..core.client_wrapper import AsyncClientWrapper, SyncClientWrapper
 from ..core.request_options import RequestOptions
 from ..types.input_audio_codec import InputAudioCodec
+from ..types.mode import Mode
 from ..types.speech_to_text_language import SpeechToTextLanguage
 from ..types.speech_to_text_model import SpeechToTextModel
 from ..types.speech_to_text_response import SpeechToTextResponse
@@ -37,6 +38,7 @@ class SpeechToTextClient:
         *,
         file: core.File,
         model: typing.Optional[SpeechToTextModel] = OMIT,
+        mode: typing.Optional[Mode] = OMIT,
         language_code: typing.Optional[SpeechToTextLanguage] = OMIT,
         input_audio_codec: typing.Optional[InputAudioCodec] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
@@ -63,12 +65,49 @@ class SpeechToTextClient:
 
         model : typing.Optional[SpeechToTextModel]
             Specifies the model to use for speech-to-text conversion.
-            Note:- Default model is `saarika:v2.5`
+
+            - **saarika:v2.5** (default): Transcribes audio in the spoken language.
+
+            - **saaras:v3**: State-of-the-art model with flexible output formats. Supports multiple modes via the `mode` parameter: transcribe, translate, verbatim, translit, codemix.
+
+        mode : typing.Optional[Mode]
+            Mode of operation. **Only applicable when using saaras:v3 model.**
+
+            Example audio: 'मेरा फोन नंबर है 9840950950'
+
+            - **transcribe** (default): Standard transcription in the original language with proper formatting and number normalization.
+              - Output: `मेरा फोन नंबर है 9840950950`
+
+            - **translate**: Translates speech from any supported Indic language to English.
+              - Output: `My phone number is 9840950950`
+
+            - **verbatim**: Exact word-for-word transcription without normalization, preserving filler words and spoken numbers as-is.
+              - Output: `मेरा फोन नंबर है नौ आठ चार zero नौ पांच zero नौ पांच zero`
+
+            - **translit**: Romanization - Transliterates speech to Latin/Roman script only.
+              - Output: `mera phone number hai 9840950950`
+
+            - **codemix**: Code-mixed text with English words in English and Indic words in native script.
+              - Output: `मेरा phone number है 9840950950`
 
         language_code : typing.Optional[SpeechToTextLanguage]
-            Specifies the language of the input audio.
-             For the `saarika:v2.5` model, it is optional.
-            `unknown`: Use this when the language is not known; the API will detect it automatically.
+            Specifies the language of the input audio in BCP-47 format.
+
+            **Note:** This parameter is optional for `saarika:v2.5` model.
+
+            **Available Options:**
+            - `unknown`: Use when the language is not known; the API will auto-detect.
+            - `hi-IN`: Hindi
+            - `bn-IN`: Bengali
+            - `kn-IN`: Kannada
+            - `ml-IN`: Malayalam
+            - `mr-IN`: Marathi
+            - `od-IN`: Odia
+            - `pa-IN`: Punjabi
+            - `ta-IN`: Tamil
+            - `te-IN`: Telugu
+            - `en-IN`: English
+            - `gu-IN`: Gujarati
 
         input_audio_codec : typing.Optional[InputAudioCodec]
             Input Audio codec/format of the input file. PCM files are supported only at 16kHz sample rate.
@@ -93,6 +132,7 @@ class SpeechToTextClient:
         _response = self._raw_client.transcribe(
             file=file,
             model=model,
+            mode=mode,
             language_code=language_code,
             input_audio_codec=input_audio_codec,
             request_options=request_options,
@@ -132,7 +172,10 @@ class SpeechToTextClient:
             Conversation context can be passed as a prompt to boost model accuracy. However, the current system is at an experimentation stage and doesn't match the prompt performance of large language models.
 
         model : typing.Optional[SpeechToTextTranslateModel]
-            Model to be used for converting speech to text in target language
+            Model to be used for speech to text translation.
+
+            - **saaras:v2.5** (default): Translation model that translates audio from any spoken Indic language to English.
+              - Example: Hindi audio → English text output
 
         input_audio_codec : typing.Optional[InputAudioCodec]
             Audio codec/format of the input file. Our API automatically detects all codec formats, but for PCM files specifically (pcm_s16le, pcm_l16, pcm_raw), you must pass this parameter. PCM files are supported only at 16kHz sample rate.
@@ -180,6 +223,7 @@ class AsyncSpeechToTextClient:
         *,
         file: core.File,
         model: typing.Optional[SpeechToTextModel] = OMIT,
+        mode: typing.Optional[Mode] = OMIT,
         language_code: typing.Optional[SpeechToTextLanguage] = OMIT,
         input_audio_codec: typing.Optional[InputAudioCodec] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
@@ -206,12 +250,49 @@ class AsyncSpeechToTextClient:
 
         model : typing.Optional[SpeechToTextModel]
             Specifies the model to use for speech-to-text conversion.
-            Note:- Default model is `saarika:v2.5`
+
+            - **saarika:v2.5** (default): Transcribes audio in the spoken language.
+
+            - **saaras:v3**: State-of-the-art model with flexible output formats. Supports multiple modes via the `mode` parameter: transcribe, translate, verbatim, translit, codemix.
+
+        mode : typing.Optional[Mode]
+            Mode of operation. **Only applicable when using saaras:v3 model.**
+
+            Example audio: 'मेरा फोन नंबर है 9840950950'
+
+            - **transcribe** (default): Standard transcription in the original language with proper formatting and number normalization.
+              - Output: `मेरा फोन नंबर है 9840950950`
+
+            - **translate**: Translates speech from any supported Indic language to English.
+              - Output: `My phone number is 9840950950`
+
+            - **verbatim**: Exact word-for-word transcription without normalization, preserving filler words and spoken numbers as-is.
+              - Output: `मेरा फोन नंबर है नौ आठ चार zero नौ पांच zero नौ पांच zero`
+
+            - **translit**: Romanization - Transliterates speech to Latin/Roman script only.
+              - Output: `mera phone number hai 9840950950`
+
+            - **codemix**: Code-mixed text with English words in English and Indic words in native script.
+              - Output: `मेरा phone number है 9840950950`
 
         language_code : typing.Optional[SpeechToTextLanguage]
-            Specifies the language of the input audio.
-             For the `saarika:v2.5` model, it is optional.
-            `unknown`: Use this when the language is not known; the API will detect it automatically.
+            Specifies the language of the input audio in BCP-47 format.
+
+            **Note:** This parameter is optional for `saarika:v2.5` model.
+
+            **Available Options:**
+            - `unknown`: Use when the language is not known; the API will auto-detect.
+            - `hi-IN`: Hindi
+            - `bn-IN`: Bengali
+            - `kn-IN`: Kannada
+            - `ml-IN`: Malayalam
+            - `mr-IN`: Marathi
+            - `od-IN`: Odia
+            - `pa-IN`: Punjabi
+            - `ta-IN`: Tamil
+            - `te-IN`: Telugu
+            - `en-IN`: English
+            - `gu-IN`: Gujarati
 
         input_audio_codec : typing.Optional[InputAudioCodec]
             Input Audio codec/format of the input file. PCM files are supported only at 16kHz sample rate.
@@ -244,6 +325,7 @@ class AsyncSpeechToTextClient:
         _response = await self._raw_client.transcribe(
             file=file,
             model=model,
+            mode=mode,
             language_code=language_code,
             input_audio_codec=input_audio_codec,
             request_options=request_options,
@@ -283,7 +365,10 @@ class AsyncSpeechToTextClient:
             Conversation context can be passed as a prompt to boost model accuracy. However, the current system is at an experimentation stage and doesn't match the prompt performance of large language models.
 
         model : typing.Optional[SpeechToTextTranslateModel]
-            Model to be used for converting speech to text in target language
+            Model to be used for speech to text translation.
+
+            - **saaras:v2.5** (default): Translation model that translates audio from any spoken Indic language to English.
+              - Example: Hindi audio → English text output
 
         input_audio_codec : typing.Optional[InputAudioCodec]
             Audio codec/format of the input file. Our API automatically detects all codec formats, but for PCM files specifically (pcm_s16le, pcm_l16, pcm_raw), you must pass this parameter. PCM files are supported only at 16kHz sample rate.

sarvamai/speech_to_text/raw_client.py

@@ -1,6 +1,5 @@
 # This file was auto-generated by Fern from our API Definition.
 
-import json
 import typing
 from json.decoder import JSONDecodeError
 
@@ -8,7 +7,6 @@ from .. import core
 from ..core.api_error import ApiError
 from ..core.client_wrapper import AsyncClientWrapper, SyncClientWrapper
 from ..core.http_response import AsyncHttpResponse, HttpResponse
-from ..core.jsonable_encoder import jsonable_encoder
 from ..core.pydantic_utilities import parse_obj_as
 from ..core.request_options import RequestOptions
 from ..errors.bad_request_error import BadRequestError
@@ -18,6 +16,7 @@ from ..errors.service_unavailable_error import ServiceUnavailableError
 from ..errors.too_many_requests_error import TooManyRequestsError
 from ..errors.unprocessable_entity_error import UnprocessableEntityError
 from ..types.input_audio_codec import InputAudioCodec
+from ..types.mode import Mode
 from ..types.speech_to_text_language import SpeechToTextLanguage
 from ..types.speech_to_text_model import SpeechToTextModel
 from ..types.speech_to_text_response import SpeechToTextResponse
@@ -37,6 +36,7 @@ class RawSpeechToTextClient:
         *,
         file: core.File,
         model: typing.Optional[SpeechToTextModel] = OMIT,
+        mode: typing.Optional[Mode] = OMIT,
         language_code: typing.Optional[SpeechToTextLanguage] = OMIT,
         input_audio_codec: typing.Optional[InputAudioCodec] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
@@ -63,12 +63,49 @@ class RawSpeechToTextClient:
 
         model : typing.Optional[SpeechToTextModel]
             Specifies the model to use for speech-to-text conversion.
-            Note:- Default model is `saarika:v2.5`
+
+            - **saarika:v2.5** (default): Transcribes audio in the spoken language.
+
+            - **saaras:v3**: State-of-the-art model with flexible output formats. Supports multiple modes via the `mode` parameter: transcribe, translate, verbatim, translit, codemix.
+
+        mode : typing.Optional[Mode]
+            Mode of operation. **Only applicable when using saaras:v3 model.**
+
+            Example audio: 'मेरा फोन नंबर है 9840950950'
+
+            - **transcribe** (default): Standard transcription in the original language with proper formatting and number normalization.
+              - Output: `मेरा फोन नंबर है 9840950950`
+
+            - **translate**: Translates speech from any supported Indic language to English.
+              - Output: `My phone number is 9840950950`
+
+            - **verbatim**: Exact word-for-word transcription without normalization, preserving filler words and spoken numbers as-is.
+              - Output: `मेरा फोन नंबर है नौ आठ चार zero नौ पांच zero नौ पांच zero`
+
+            - **translit**: Romanization - Transliterates speech to Latin/Roman script only.
+              - Output: `mera phone number hai 9840950950`
+
+            - **codemix**: Code-mixed text with English words in English and Indic words in native script.
+              - Output: `मेरा phone number है 9840950950`
 
         language_code : typing.Optional[SpeechToTextLanguage]
-            Specifies the language of the input audio.
-             For the `saarika:v2.5` model, it is optional.
-            `unknown`: Use this when the language is not known; the API will detect it automatically.
+            Specifies the language of the input audio in BCP-47 format.
+
+            **Note:** This parameter is optional for `saarika:v2.5` model.
+
+            **Available Options:**
+            - `unknown`: Use when the language is not known; the API will auto-detect.
+            - `hi-IN`: Hindi
+            - `bn-IN`: Bengali
+            - `kn-IN`: Kannada
+            - `ml-IN`: Malayalam
+            - `mr-IN`: Marathi
+            - `od-IN`: Odia
+            - `pa-IN`: Punjabi
+            - `ta-IN`: Tamil
+            - `te-IN`: Telugu
+            - `en-IN`: English
+            - `gu-IN`: Gujarati
 
         input_audio_codec : typing.Optional[InputAudioCodec]
             Input Audio codec/format of the input file. PCM files are supported only at 16kHz sample rate.
@@ -86,7 +123,8 @@ class RawSpeechToTextClient:
             base_url=self._client_wrapper.get_environment().base,
             method="POST",
             data={
-                "model": json.dumps(jsonable_encoder(model)),
+                "model": model,
+                "mode": mode,
                 "language_code": language_code,
                 "input_audio_codec": input_audio_codec,
             },
@@ -111,9 +149,9 @@ class RawSpeechToTextClient:
                 raise BadRequestError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -122,9 +160,9 @@ class RawSpeechToTextClient:
                 raise ForbiddenError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -133,9 +171,9 @@ class RawSpeechToTextClient:
                 raise UnprocessableEntityError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -144,9 +182,9 @@ class RawSpeechToTextClient:
                 raise TooManyRequestsError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -155,9 +193,9 @@ class RawSpeechToTextClient:
                 raise InternalServerError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -166,9 +204,9 @@ class RawSpeechToTextClient:
                 raise ServiceUnavailableError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -211,7 +249,10 @@ class RawSpeechToTextClient:
             Conversation context can be passed as a prompt to boost model accuracy. However, the current system is at an experimentation stage and doesn't match the prompt performance of large language models.
 
         model : typing.Optional[SpeechToTextTranslateModel]
-            Model to be used for converting speech to text in target language
+            Model to be used for speech to text translation.
+
+            - **saaras:v2.5** (default): Translation model that translates audio from any spoken Indic language to English.
+              - Example: Hindi audio → English text output
 
         input_audio_codec : typing.Optional[InputAudioCodec]
             Audio codec/format of the input file. Our API automatically detects all codec formats, but for PCM files specifically (pcm_s16le, pcm_l16, pcm_raw), you must pass this parameter. PCM files are supported only at 16kHz sample rate.
@@ -230,7 +271,7 @@ class RawSpeechToTextClient:
             method="POST",
             data={
                 "prompt": prompt,
-                "model": json.dumps(jsonable_encoder(model)),
+                "model": model,
                 "input_audio_codec": input_audio_codec,
             },
             files={
@@ -254,9 +295,9 @@ class RawSpeechToTextClient:
                 raise BadRequestError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -265,9 +306,9 @@ class RawSpeechToTextClient:
                 raise ForbiddenError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -276,9 +317,9 @@ class RawSpeechToTextClient:
                 raise UnprocessableEntityError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -287,9 +328,9 @@ class RawSpeechToTextClient:
                 raise TooManyRequestsError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -298,9 +339,9 @@ class RawSpeechToTextClient:
                 raise InternalServerError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -309,9 +350,9 @@ class RawSpeechToTextClient:
                 raise ServiceUnavailableError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -331,6 +372,7 @@ class AsyncRawSpeechToTextClient:
         *,
         file: core.File,
         model: typing.Optional[SpeechToTextModel] = OMIT,
+        mode: typing.Optional[Mode] = OMIT,
         language_code: typing.Optional[SpeechToTextLanguage] = OMIT,
         input_audio_codec: typing.Optional[InputAudioCodec] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
@@ -357,12 +399,49 @@ class AsyncRawSpeechToTextClient:
 
         model : typing.Optional[SpeechToTextModel]
             Specifies the model to use for speech-to-text conversion.
-            Note:- Default model is `saarika:v2.5`
+
+            - **saarika:v2.5** (default): Transcribes audio in the spoken language.
+
+            - **saaras:v3**: State-of-the-art model with flexible output formats. Supports multiple modes via the `mode` parameter: transcribe, translate, verbatim, translit, codemix.
+
+        mode : typing.Optional[Mode]
+            Mode of operation. **Only applicable when using saaras:v3 model.**
+
+            Example audio: 'मेरा फोन नंबर है 9840950950'
+
+            - **transcribe** (default): Standard transcription in the original language with proper formatting and number normalization.
+              - Output: `मेरा फोन नंबर है 9840950950`
+
+            - **translate**: Translates speech from any supported Indic language to English.
+              - Output: `My phone number is 9840950950`
+
+            - **verbatim**: Exact word-for-word transcription without normalization, preserving filler words and spoken numbers as-is.
+              - Output: `मेरा फोन नंबर है नौ आठ चार zero नौ पांच zero नौ पांच zero`
+
+            - **translit**: Romanization - Transliterates speech to Latin/Roman script only.
+              - Output: `mera phone number hai 9840950950`
+
+            - **codemix**: Code-mixed text with English words in English and Indic words in native script.
+              - Output: `मेरा phone number है 9840950950`
 
         language_code : typing.Optional[SpeechToTextLanguage]
-            Specifies the language of the input audio.
-             For the `saarika:v2.5` model, it is optional.
-            `unknown`: Use this when the language is not known; the API will detect it automatically.
+            Specifies the language of the input audio in BCP-47 format.
+
+            **Note:** This parameter is optional for `saarika:v2.5` model.
+
+            **Available Options:**
+            - `unknown`: Use when the language is not known; the API will auto-detect.
+            - `hi-IN`: Hindi
+            - `bn-IN`: Bengali
+            - `kn-IN`: Kannada
+            - `ml-IN`: Malayalam
+            - `mr-IN`: Marathi
+            - `od-IN`: Odia
+            - `pa-IN`: Punjabi
+            - `ta-IN`: Tamil
+            - `te-IN`: Telugu
+            - `en-IN`: English
+            - `gu-IN`: Gujarati
 
         input_audio_codec : typing.Optional[InputAudioCodec]
             Input Audio codec/format of the input file. PCM files are supported only at 16kHz sample rate.
@@ -380,7 +459,8 @@ class AsyncRawSpeechToTextClient:
             base_url=self._client_wrapper.get_environment().base,
             method="POST",
             data={
-                "model": json.dumps(jsonable_encoder(model)),
+                "model": model,
+                "mode": mode,
                 "language_code": language_code,
                 "input_audio_codec": input_audio_codec,
             },
@@ -405,9 +485,9 @@ class AsyncRawSpeechToTextClient:
                 raise BadRequestError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -416,9 +496,9 @@ class AsyncRawSpeechToTextClient:
                 raise ForbiddenError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -427,9 +507,9 @@ class AsyncRawSpeechToTextClient:
                 raise UnprocessableEntityError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -438,9 +518,9 @@ class AsyncRawSpeechToTextClient:
                 raise TooManyRequestsError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -449,9 +529,9 @@ class AsyncRawSpeechToTextClient:
                 raise InternalServerError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -460,9 +540,9 @@ class AsyncRawSpeechToTextClient:
                 raise ServiceUnavailableError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -505,7 +585,10 @@ class AsyncRawSpeechToTextClient:
             Conversation context can be passed as a prompt to boost model accuracy. However, the current system is at an experimentation stage and doesn't match the prompt performance of large language models.
 
         model : typing.Optional[SpeechToTextTranslateModel]
-            Model to be used for converting speech to text in target language
+            Model to be used for speech to text translation.
+
+            - **saaras:v2.5** (default): Translation model that translates audio from any spoken Indic language to English.
+              - Example: Hindi audio → English text output
 
         input_audio_codec : typing.Optional[InputAudioCodec]
             Audio codec/format of the input file. Our API automatically detects all codec formats, but for PCM files specifically (pcm_s16le, pcm_l16, pcm_raw), you must pass this parameter. PCM files are supported only at 16kHz sample rate.
@@ -524,7 +607,7 @@ class AsyncRawSpeechToTextClient:
             method="POST",
             data={
                 "prompt": prompt,
-                "model": json.dumps(jsonable_encoder(model)),
+                "model": model,
                 "input_audio_codec": input_audio_codec,
             },
             files={
@@ -548,9 +631,9 @@ class AsyncRawSpeechToTextClient:
                 raise BadRequestError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -559,9 +642,9 @@ class AsyncRawSpeechToTextClient:
                 raise ForbiddenError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -570,9 +653,9 @@ class AsyncRawSpeechToTextClient:
                 raise UnprocessableEntityError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -581,9 +664,9 @@ class AsyncRawSpeechToTextClient:
                 raise TooManyRequestsError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -592,9 +675,9 @@ class AsyncRawSpeechToTextClient:
                 raise InternalServerError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -603,9 +686,9 @@ class AsyncRawSpeechToTextClient:
                 raise ServiceUnavailableError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Any,
+                        typing.Optional[typing.Any],
                         parse_obj_as(
-                            type_=typing.Any,  # type: ignore
+                            type_=typing.Optional[typing.Any],  # type: ignore
                             object_=_response.json(),
                         ),
                     ),