sarvamai 0.1.23a5__tar.gz → 0.1.23a6__tar.gz

{sarvamai-0.1.23a5 → sarvamai-0.1.23a6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sarvamai
-Version: 0.1.23a5
+Version: 0.1.23a6
 Summary: 
 Requires-Python: >=3.8,<4.0
 Classifier: Intended Audience :: Developers

{sarvamai-0.1.23a5 → sarvamai-0.1.23a6}/pyproject.toml

@@ -3,7 +3,7 @@ name = "sarvamai"
 
 [tool.poetry]
 name = "sarvamai"
-version = "0.1.23a5"
+version = "0.1.23a6"
 description = ""
 readme = "README.md"
 authors = []

{sarvamai-0.1.23a5 → sarvamai-0.1.23a6}/src/sarvamai/__init__.py

@@ -188,6 +188,7 @@ from .requests import (
 from .speech_to_text_streaming import (
     SpeechToTextStreamingFlushSignal,
     SpeechToTextStreamingHighVadSensitivity,
+    SpeechToTextStreamingInputAudioCodec,
     SpeechToTextStreamingLanguageCode,
     SpeechToTextStreamingMode,
     SpeechToTextStreamingModel,
@@ -196,6 +197,7 @@ from .speech_to_text_streaming import (
 from .speech_to_text_translate_streaming import (
     SpeechToTextTranslateStreamingFlushSignal,
     SpeechToTextTranslateStreamingHighVadSensitivity,
+    SpeechToTextTranslateStreamingInputAudioCodec,
     SpeechToTextTranslateStreamingVadSignals,
 )
 from .text_to_speech_streaming import TextToSpeechStreamingModel, TextToSpeechStreamingSendCompletionEvent
@@ -320,6 +322,7 @@ __all__ = [
     "SpeechToTextResponseParams",
     "SpeechToTextStreamingFlushSignal",
     "SpeechToTextStreamingHighVadSensitivity",
+    "SpeechToTextStreamingInputAudioCodec",
     "SpeechToTextStreamingLanguageCode",
     "SpeechToTextStreamingMode",
     "SpeechToTextStreamingModel",
@@ -338,6 +341,7 @@ __all__ = [
     "SpeechToTextTranslateResponseParams",
     "SpeechToTextTranslateStreamingFlushSignal",
     "SpeechToTextTranslateStreamingHighVadSensitivity",
+    "SpeechToTextTranslateStreamingInputAudioCodec",
     "SpeechToTextTranslateStreamingResponse",
     "SpeechToTextTranslateStreamingResponseParams",
     "SpeechToTextTranslateStreamingVadSignals",

{sarvamai-0.1.23a5 → sarvamai-0.1.23a6}/src/sarvamai/core/client_wrapper.py

@@ -23,10 +23,10 @@ class BaseClientWrapper:
 
     def get_headers(self) -> typing.Dict[str, str]:
         headers: typing.Dict[str, str] = {
-            "User-Agent": "sarvamai/0.1.23a5",
+            "User-Agent": "sarvamai/0.1.23a6",
             "X-Fern-Language": "Python",
             "X-Fern-SDK-Name": "sarvamai",
-            "X-Fern-SDK-Version": "0.1.23a5",
+            "X-Fern-SDK-Version": "0.1.23a6",
             **(self.get_custom_headers() or {}),
         }
         headers["api-subscription-key"] = self.api_subscription_key

sarvamai-0.1.23a6/src/sarvamai/requests/speech_to_text_response.py

@@ -0,0 +1,20 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing_extensions
+
+
+class SpeechToTextResponseParams(typing_extensions.TypedDict):
+    request_id: str
+    """
+    Unique identifier for the request
+    """
+
+    transcript: str
+    """
+    The transcribed text from the provided audio file.
+    """
+
+    language_code: str
+    """
+    The BCP-47 code of language spoken in the input (e.g., hi-IN, en-IN). If multiple languages are detected, returns the most predominant spoken language.
+    """

{sarvamai-0.1.23a5 → sarvamai-0.1.23a6}/src/sarvamai/requests/speech_to_text_transcription_data.py

@@ -32,4 +32,18 @@ class SpeechToTextTranscriptionDataParams(typing_extensions.TypedDict):
     BCP-47 code of detected language
     """
 
+    language_probability: typing_extensions.NotRequired[float]
+    """
+    Float value (0.0 to 1.0) indicating the probability of the detected language being correct. Higher values indicate higher confidence.
+    
+    **When it returns a value:**
+    - When `language_code` is not provided in the request
+    - When `language_code` is set to `unknown`
+    
+    **When it returns null:**
+    - When a specific `language_code` is provided (language detection is skipped)
+    
+    The parameter is always present in the response.
+    """
+
     metrics: TranscriptionMetricsParams

sarvamai-0.1.23a6/src/sarvamai/requests/speech_to_text_translate_response.py

@@ -0,0 +1,20 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing_extensions
+
+
+class SpeechToTextTranslateResponseParams(typing_extensions.TypedDict):
+    request_id: str
+    """
+    Unique identifier for the request
+    """
+
+    transcript: str
+    """
+    English translation of the provided speech
+    """
+
+    language_code: str
+    """
+    The BCP-47 code of the detected source language spoken in the input (e.g., hi-IN, kn-IN).
+    """

{sarvamai-0.1.23a5 → sarvamai-0.1.23a6}/src/sarvamai/requests/speech_to_text_translate_transcription_data.py

@@ -20,4 +20,17 @@ class SpeechToTextTranslateTranscriptionDataParams(typing_extensions.TypedDict):
     BCP-47 code of detected source language (null when language detection is in progress)
     """
 
+    language_probability: typing_extensions.NotRequired[float]
+    """
+    Float value (0.0 to 1.0) indicating the probability of the detected source language being correct. Higher values indicate higher confidence.
+    
+    **When it returns a value:**
+    - Always returns a value as source language is auto-detected for translation
+    
+    **When it returns null:**
+    - When language detection confidence is unavailable
+    
+    The parameter is always present in the response.
+    """
+
     metrics: TranscriptionMetricsParams

{sarvamai-0.1.23a5 → sarvamai-0.1.23a6}/src/sarvamai/speech_to_text_streaming/__init__.py

@@ -5,6 +5,7 @@
 from .types import (
     SpeechToTextStreamingFlushSignal,
     SpeechToTextStreamingHighVadSensitivity,
+    SpeechToTextStreamingInputAudioCodec,
     SpeechToTextStreamingLanguageCode,
     SpeechToTextStreamingMode,
     SpeechToTextStreamingModel,
@@ -14,6 +15,7 @@ from .types import (
 __all__ = [
     "SpeechToTextStreamingFlushSignal",
     "SpeechToTextStreamingHighVadSensitivity",
+    "SpeechToTextStreamingInputAudioCodec",
     "SpeechToTextStreamingLanguageCode",
     "SpeechToTextStreamingMode",
     "SpeechToTextStreamingModel",

{sarvamai-0.1.23a5 → sarvamai-0.1.23a6}/src/sarvamai/speech_to_text_streaming/client.py

@@ -13,6 +13,7 @@ from .raw_client import AsyncRawSpeechToTextStreamingClient, RawSpeechToTextStre
 from .socket_client import AsyncSpeechToTextStreamingSocketClient, SpeechToTextStreamingSocketClient
 from .types.speech_to_text_streaming_flush_signal import SpeechToTextStreamingFlushSignal
 from .types.speech_to_text_streaming_high_vad_sensitivity import SpeechToTextStreamingHighVadSensitivity
+from .types.speech_to_text_streaming_input_audio_codec import SpeechToTextStreamingInputAudioCodec
 from .types.speech_to_text_streaming_language_code import SpeechToTextStreamingLanguageCode
 from .types.speech_to_text_streaming_mode import SpeechToTextStreamingMode
 from .types.speech_to_text_streaming_model import SpeechToTextStreamingModel
@@ -50,6 +51,7 @@ class SpeechToTextStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextStreamingFlushSignal] = None,
+        input_audio_codec: typing.Optional[SpeechToTextStreamingInputAudioCodec] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[SpeechToTextStreamingSocketClient]:
@@ -65,7 +67,7 @@ class SpeechToTextStreamingClient:
         language_code : SpeechToTextStreamingLanguageCode
             Specifies the language of the input audio in BCP-47 format.
 
-            **Available Options:**
+            **Available Options (saarika:v2.5):**
             - `hi-IN`: Hindi
             - `bn-IN`: Bengali
             - `gu-IN`: Gujarati
@@ -78,6 +80,20 @@ class SpeechToTextStreamingClient:
             - `te-IN`: Telugu
             - `en-IN`: English
 
+            **Additional Options (saaras:v3 only):**
+            - `as-IN`: Assamese
+            - `ur-IN`: Urdu
+            - `ne-IN`: Nepali
+            - `kok-IN`: Konkani
+            - `ks-IN`: Kashmiri
+            - `sd-IN`: Sindhi
+            - `sa-IN`: Sanskrit
+            - `sat-IN`: Santali
+            - `mni-IN`: Manipuri
+            - `brx-IN`: Bodo
+            - `mai-IN`: Maithili
+            - `doi-IN`: Dogri
+
         model : typing.Optional[SpeechToTextStreamingModel]
             Specifies the model to use for speech-to-text conversion.
 
@@ -117,6 +133,10 @@ class SpeechToTextStreamingClient:
         flush_signal : typing.Optional[SpeechToTextStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription
 
+        input_audio_codec : typing.Optional[SpeechToTextStreamingInputAudioCodec]
+            Audio codec/format of the input stream. Use this when sending raw PCM audio.
+            Supported values: wav, pcm_s16le, pcm_l16, pcm_raw.
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -143,6 +163,8 @@ class SpeechToTextStreamingClient:
             query_params = query_params.add("vad_signals", vad_signals)
         if flush_signal is not None:
             query_params = query_params.add("flush_signal", flush_signal)
+        if input_audio_codec is not None:
+            query_params = query_params.add("input_audio_codec", input_audio_codec)
         ws_url = ws_url + f"?{query_params}"
         headers = self._raw_client._client_wrapper.get_headers()
         if api_subscription_key is not None:
@@ -193,6 +215,7 @@ class AsyncSpeechToTextStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextStreamingFlushSignal] = None,
+        input_audio_codec: typing.Optional[SpeechToTextStreamingInputAudioCodec] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncSpeechToTextStreamingSocketClient]:
@@ -208,7 +231,7 @@ class AsyncSpeechToTextStreamingClient:
         language_code : SpeechToTextStreamingLanguageCode
             Specifies the language of the input audio in BCP-47 format.
 
-            **Available Options:**
+            **Available Options (saarika:v2.5):**
             - `hi-IN`: Hindi
             - `bn-IN`: Bengali
             - `gu-IN`: Gujarati
@@ -221,6 +244,20 @@ class AsyncSpeechToTextStreamingClient:
             - `te-IN`: Telugu
             - `en-IN`: English
 
+            **Additional Options (saaras:v3 only):**
+            - `as-IN`: Assamese
+            - `ur-IN`: Urdu
+            - `ne-IN`: Nepali
+            - `kok-IN`: Konkani
+            - `ks-IN`: Kashmiri
+            - `sd-IN`: Sindhi
+            - `sa-IN`: Sanskrit
+            - `sat-IN`: Santali
+            - `mni-IN`: Manipuri
+            - `brx-IN`: Bodo
+            - `mai-IN`: Maithili
+            - `doi-IN`: Dogri
+
         model : typing.Optional[SpeechToTextStreamingModel]
             Specifies the model to use for speech-to-text conversion.
 
@@ -260,6 +297,10 @@ class AsyncSpeechToTextStreamingClient:
         flush_signal : typing.Optional[SpeechToTextStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription
 
+        input_audio_codec : typing.Optional[SpeechToTextStreamingInputAudioCodec]
+            Audio codec/format of the input stream. Use this when sending raw PCM audio.
+            Supported values: wav, pcm_s16le, pcm_l16, pcm_raw.
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -286,6 +327,8 @@ class AsyncSpeechToTextStreamingClient:
             query_params = query_params.add("vad_signals", vad_signals)
         if flush_signal is not None:
             query_params = query_params.add("flush_signal", flush_signal)
+        if input_audio_codec is not None:
+            query_params = query_params.add("input_audio_codec", input_audio_codec)
         ws_url = ws_url + f"?{query_params}"
         headers = self._raw_client._client_wrapper.get_headers()
         if api_subscription_key is not None:

{sarvamai-0.1.23a5 → sarvamai-0.1.23a6}/src/sarvamai/speech_to_text_streaming/raw_client.py

@@ -12,6 +12,7 @@ from ..core.request_options import RequestOptions
 from .socket_client import AsyncSpeechToTextStreamingSocketClient, SpeechToTextStreamingSocketClient
 from .types.speech_to_text_streaming_flush_signal import SpeechToTextStreamingFlushSignal
 from .types.speech_to_text_streaming_high_vad_sensitivity import SpeechToTextStreamingHighVadSensitivity
+from .types.speech_to_text_streaming_input_audio_codec import SpeechToTextStreamingInputAudioCodec
 from .types.speech_to_text_streaming_language_code import SpeechToTextStreamingLanguageCode
 from .types.speech_to_text_streaming_mode import SpeechToTextStreamingMode
 from .types.speech_to_text_streaming_model import SpeechToTextStreamingModel
@@ -38,6 +39,7 @@ class RawSpeechToTextStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextStreamingFlushSignal] = None,
+        input_audio_codec: typing.Optional[SpeechToTextStreamingInputAudioCodec] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[SpeechToTextStreamingSocketClient]:
@@ -53,7 +55,7 @@ class RawSpeechToTextStreamingClient:
         language_code : SpeechToTextStreamingLanguageCode
             Specifies the language of the input audio in BCP-47 format.
 
-            **Available Options:**
+            **Available Options (saarika:v2.5):**
             - `hi-IN`: Hindi
             - `bn-IN`: Bengali
             - `gu-IN`: Gujarati
@@ -66,6 +68,20 @@ class RawSpeechToTextStreamingClient:
             - `te-IN`: Telugu
             - `en-IN`: English
 
+            **Additional Options (saaras:v3 only):**
+            - `as-IN`: Assamese
+            - `ur-IN`: Urdu
+            - `ne-IN`: Nepali
+            - `kok-IN`: Konkani
+            - `ks-IN`: Kashmiri
+            - `sd-IN`: Sindhi
+            - `sa-IN`: Sanskrit
+            - `sat-IN`: Santali
+            - `mni-IN`: Manipuri
+            - `brx-IN`: Bodo
+            - `mai-IN`: Maithili
+            - `doi-IN`: Dogri
+
         model : typing.Optional[SpeechToTextStreamingModel]
             Specifies the model to use for speech-to-text conversion.
 
@@ -105,6 +121,10 @@ class RawSpeechToTextStreamingClient:
         flush_signal : typing.Optional[SpeechToTextStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription
 
+        input_audio_codec : typing.Optional[SpeechToTextStreamingInputAudioCodec]
+            Audio codec/format of the input stream. Use this when sending raw PCM audio.
+            Supported values: wav, pcm_s16le, pcm_l16, pcm_raw.
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -131,6 +151,8 @@ class RawSpeechToTextStreamingClient:
             query_params = query_params.add("vad_signals", vad_signals)
         if flush_signal is not None:
             query_params = query_params.add("flush_signal", flush_signal)
+        if input_audio_codec is not None:
+            query_params = query_params.add("input_audio_codec", input_audio_codec)
         ws_url = ws_url + f"?{query_params}"
         headers = self._client_wrapper.get_headers()
         if api_subscription_key is not None:
@@ -170,6 +192,7 @@ class AsyncRawSpeechToTextStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextStreamingFlushSignal] = None,
+        input_audio_codec: typing.Optional[SpeechToTextStreamingInputAudioCodec] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncSpeechToTextStreamingSocketClient]:
@@ -185,7 +208,7 @@ class AsyncRawSpeechToTextStreamingClient:
         language_code : SpeechToTextStreamingLanguageCode
             Specifies the language of the input audio in BCP-47 format.
 
-            **Available Options:**
+            **Available Options (saarika:v2.5):**
             - `hi-IN`: Hindi
             - `bn-IN`: Bengali
             - `gu-IN`: Gujarati
@@ -198,6 +221,20 @@ class AsyncRawSpeechToTextStreamingClient:
             - `te-IN`: Telugu
             - `en-IN`: English
 
+            **Additional Options (saaras:v3 only):**
+            - `as-IN`: Assamese
+            - `ur-IN`: Urdu
+            - `ne-IN`: Nepali
+            - `kok-IN`: Konkani
+            - `ks-IN`: Kashmiri
+            - `sd-IN`: Sindhi
+            - `sa-IN`: Sanskrit
+            - `sat-IN`: Santali
+            - `mni-IN`: Manipuri
+            - `brx-IN`: Bodo
+            - `mai-IN`: Maithili
+            - `doi-IN`: Dogri
+
         model : typing.Optional[SpeechToTextStreamingModel]
             Specifies the model to use for speech-to-text conversion.
 
@@ -237,6 +274,10 @@ class AsyncRawSpeechToTextStreamingClient:
         flush_signal : typing.Optional[SpeechToTextStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription
 
+        input_audio_codec : typing.Optional[SpeechToTextStreamingInputAudioCodec]
+            Audio codec/format of the input stream. Use this when sending raw PCM audio.
+            Supported values: wav, pcm_s16le, pcm_l16, pcm_raw.
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -263,6 +304,8 @@ class AsyncRawSpeechToTextStreamingClient:
             query_params = query_params.add("vad_signals", vad_signals)
         if flush_signal is not None:
             query_params = query_params.add("flush_signal", flush_signal)
+        if input_audio_codec is not None:
+            query_params = query_params.add("input_audio_codec", input_audio_codec)
         ws_url = ws_url + f"?{query_params}"
         headers = self._client_wrapper.get_headers()
         if api_subscription_key is not None:

{sarvamai-0.1.23a5 → sarvamai-0.1.23a6}/src/sarvamai/speech_to_text_streaming/types/__init__.py

@@ -4,6 +4,7 @@
 
 from .speech_to_text_streaming_flush_signal import SpeechToTextStreamingFlushSignal
 from .speech_to_text_streaming_high_vad_sensitivity import SpeechToTextStreamingHighVadSensitivity
+from .speech_to_text_streaming_input_audio_codec import SpeechToTextStreamingInputAudioCodec
 from .speech_to_text_streaming_language_code import SpeechToTextStreamingLanguageCode
 from .speech_to_text_streaming_mode import SpeechToTextStreamingMode
 from .speech_to_text_streaming_model import SpeechToTextStreamingModel
@@ -12,6 +13,7 @@ from .speech_to_text_streaming_vad_signals import SpeechToTextStreamingVadSignal
 __all__ = [
     "SpeechToTextStreamingFlushSignal",
     "SpeechToTextStreamingHighVadSensitivity",
+    "SpeechToTextStreamingInputAudioCodec",
     "SpeechToTextStreamingLanguageCode",
     "SpeechToTextStreamingMode",
     "SpeechToTextStreamingModel",

sarvamai-0.1.23a6/src/sarvamai/speech_to_text_streaming/types/speech_to_text_streaming_input_audio_codec.py

@@ -0,0 +1,7 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+SpeechToTextStreamingInputAudioCodec = typing.Union[
+    typing.Literal["wav", "pcm_s16le", "pcm_l16", "pcm_raw"], typing.Any
+]

sarvamai-0.1.23a6/src/sarvamai/speech_to_text_streaming/types/speech_to_text_streaming_language_code.py

@@ -0,0 +1,32 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+SpeechToTextStreamingLanguageCode = typing.Union[
+    typing.Literal[
+        "en-IN",
+        "hi-IN",
+        "bn-IN",
+        "gu-IN",
+        "kn-IN",
+        "ml-IN",
+        "mr-IN",
+        "od-IN",
+        "pa-IN",
+        "ta-IN",
+        "te-IN",
+        "as-IN",
+        "ur-IN",
+        "ne-IN",
+        "kok-IN",
+        "ks-IN",
+        "sd-IN",
+        "sa-IN",
+        "sat-IN",
+        "mni-IN",
+        "brx-IN",
+        "mai-IN",
+        "doi-IN",
+    ],
+    typing.Any,
+]

{sarvamai-0.1.23a5 → sarvamai-0.1.23a6}/src/sarvamai/speech_to_text_translate_streaming/__init__.py

@@ -5,11 +5,13 @@
 from .types import (
     SpeechToTextTranslateStreamingFlushSignal,
     SpeechToTextTranslateStreamingHighVadSensitivity,
+    SpeechToTextTranslateStreamingInputAudioCodec,
     SpeechToTextTranslateStreamingVadSignals,
 )
 
 __all__ = [
     "SpeechToTextTranslateStreamingFlushSignal",
     "SpeechToTextTranslateStreamingHighVadSensitivity",
+    "SpeechToTextTranslateStreamingInputAudioCodec",
     "SpeechToTextTranslateStreamingVadSignals",
 ]

{sarvamai-0.1.23a5 → sarvamai-0.1.23a6}/src/sarvamai/speech_to_text_translate_streaming/client.py

@@ -15,6 +15,7 @@ from .types.speech_to_text_translate_streaming_flush_signal import SpeechToTextT
 from .types.speech_to_text_translate_streaming_high_vad_sensitivity import (
     SpeechToTextTranslateStreamingHighVadSensitivity,
 )
+from .types.speech_to_text_translate_streaming_input_audio_codec import SpeechToTextTranslateStreamingInputAudioCodec
 from .types.speech_to_text_translate_streaming_vad_signals import SpeechToTextTranslateStreamingVadSignals
 
 try:
@@ -47,6 +48,7 @@ class SpeechToTextTranslateStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextTranslateStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextTranslateStreamingFlushSignal] = None,
+        input_audio_codec: typing.Optional[SpeechToTextTranslateStreamingInputAudioCodec] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[SpeechToTextTranslateStreamingSocketClient]:
@@ -77,6 +79,10 @@ class SpeechToTextTranslateStreamingClient:
         flush_signal : typing.Optional[SpeechToTextTranslateStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription and translation
 
+        input_audio_codec : typing.Optional[SpeechToTextTranslateStreamingInputAudioCodec]
+            Audio codec/format of the input stream. Use this when sending raw PCM audio.
+            Supported values: wav, pcm_s16le, pcm_l16, pcm_raw.
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -99,6 +105,8 @@ class SpeechToTextTranslateStreamingClient:
             query_params = query_params.add("vad_signals", vad_signals)
         if flush_signal is not None:
             query_params = query_params.add("flush_signal", flush_signal)
+        if input_audio_codec is not None:
+            query_params = query_params.add("input_audio_codec", input_audio_codec)
         ws_url = ws_url + f"?{query_params}"
         headers = self._raw_client._client_wrapper.get_headers()
         if api_subscription_key is not None:
@@ -147,6 +155,7 @@ class AsyncSpeechToTextTranslateStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextTranslateStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextTranslateStreamingFlushSignal] = None,
+        input_audio_codec: typing.Optional[SpeechToTextTranslateStreamingInputAudioCodec] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncSpeechToTextTranslateStreamingSocketClient]:
@@ -177,6 +186,10 @@ class AsyncSpeechToTextTranslateStreamingClient:
         flush_signal : typing.Optional[SpeechToTextTranslateStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription and translation
 
+        input_audio_codec : typing.Optional[SpeechToTextTranslateStreamingInputAudioCodec]
+            Audio codec/format of the input stream. Use this when sending raw PCM audio.
+            Supported values: wav, pcm_s16le, pcm_l16, pcm_raw.
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -199,6 +212,8 @@ class AsyncSpeechToTextTranslateStreamingClient:
             query_params = query_params.add("vad_signals", vad_signals)
         if flush_signal is not None:
             query_params = query_params.add("flush_signal", flush_signal)
+        if input_audio_codec is not None:
+            query_params = query_params.add("input_audio_codec", input_audio_codec)
         ws_url = ws_url + f"?{query_params}"
         headers = self._raw_client._client_wrapper.get_headers()
         if api_subscription_key is not None:

{sarvamai-0.1.23a5 → sarvamai-0.1.23a6}/src/sarvamai/speech_to_text_translate_streaming/raw_client.py

@@ -14,6 +14,7 @@ from .types.speech_to_text_translate_streaming_flush_signal import SpeechToTextT
 from .types.speech_to_text_translate_streaming_high_vad_sensitivity import (
     SpeechToTextTranslateStreamingHighVadSensitivity,
 )
+from .types.speech_to_text_translate_streaming_input_audio_codec import SpeechToTextTranslateStreamingInputAudioCodec
 from .types.speech_to_text_translate_streaming_vad_signals import SpeechToTextTranslateStreamingVadSignals
 
 try:
@@ -35,6 +36,7 @@ class RawSpeechToTextTranslateStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextTranslateStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextTranslateStreamingFlushSignal] = None,
+        input_audio_codec: typing.Optional[SpeechToTextTranslateStreamingInputAudioCodec] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[SpeechToTextTranslateStreamingSocketClient]:
@@ -65,6 +67,10 @@ class RawSpeechToTextTranslateStreamingClient:
         flush_signal : typing.Optional[SpeechToTextTranslateStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription and translation
 
+        input_audio_codec : typing.Optional[SpeechToTextTranslateStreamingInputAudioCodec]
+            Audio codec/format of the input stream. Use this when sending raw PCM audio.
+            Supported values: wav, pcm_s16le, pcm_l16, pcm_raw.
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -87,6 +93,8 @@ class RawSpeechToTextTranslateStreamingClient:
             query_params = query_params.add("vad_signals", vad_signals)
         if flush_signal is not None:
             query_params = query_params.add("flush_signal", flush_signal)
+        if input_audio_codec is not None:
+            query_params = query_params.add("input_audio_codec", input_audio_codec)
         ws_url = ws_url + f"?{query_params}"
         headers = self._client_wrapper.get_headers()
         if api_subscription_key is not None:
@@ -124,6 +132,7 @@ class AsyncRawSpeechToTextTranslateStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextTranslateStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextTranslateStreamingFlushSignal] = None,
+        input_audio_codec: typing.Optional[SpeechToTextTranslateStreamingInputAudioCodec] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncSpeechToTextTranslateStreamingSocketClient]:
@@ -154,6 +163,10 @@ class AsyncRawSpeechToTextTranslateStreamingClient:
         flush_signal : typing.Optional[SpeechToTextTranslateStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription and translation
 
+        input_audio_codec : typing.Optional[SpeechToTextTranslateStreamingInputAudioCodec]
+            Audio codec/format of the input stream. Use this when sending raw PCM audio.
+            Supported values: wav, pcm_s16le, pcm_l16, pcm_raw.
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -176,6 +189,8 @@ class AsyncRawSpeechToTextTranslateStreamingClient:
             query_params = query_params.add("vad_signals", vad_signals)
         if flush_signal is not None:
             query_params = query_params.add("flush_signal", flush_signal)
+        if input_audio_codec is not None:
+            query_params = query_params.add("input_audio_codec", input_audio_codec)
         ws_url = ws_url + f"?{query_params}"
         headers = self._client_wrapper.get_headers()
         if api_subscription_key is not None:

{sarvamai-0.1.23a5 → sarvamai-0.1.23a6}/src/sarvamai/speech_to_text_translate_streaming/types/__init__.py

@@ -4,10 +4,12 @@
 
 from .speech_to_text_translate_streaming_flush_signal import SpeechToTextTranslateStreamingFlushSignal
 from .speech_to_text_translate_streaming_high_vad_sensitivity import SpeechToTextTranslateStreamingHighVadSensitivity
+from .speech_to_text_translate_streaming_input_audio_codec import SpeechToTextTranslateStreamingInputAudioCodec
 from .speech_to_text_translate_streaming_vad_signals import SpeechToTextTranslateStreamingVadSignals
 
 __all__ = [
     "SpeechToTextTranslateStreamingFlushSignal",
     "SpeechToTextTranslateStreamingHighVadSensitivity",
+    "SpeechToTextTranslateStreamingInputAudioCodec",
     "SpeechToTextTranslateStreamingVadSignals",
 ]

sarvamai-0.1.23a6/src/sarvamai/speech_to_text_translate_streaming/types/speech_to_text_translate_streaming_input_audio_codec.py

@@ -0,0 +1,7 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+SpeechToTextTranslateStreamingInputAudioCodec = typing.Union[
+    typing.Literal["wav", "pcm_s16le", "pcm_l16", "pcm_raw"], typing.Any
+]

sarvamai-0.1.23a6/src/sarvamai/types/speech_to_text_language.py

@@ -0,0 +1,33 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+SpeechToTextLanguage = typing.Union[
+    typing.Literal[
+        "unknown",
+        "hi-IN",
+        "bn-IN",
+        "kn-IN",
+        "ml-IN",
+        "mr-IN",
+        "od-IN",
+        "pa-IN",
+        "ta-IN",
+        "te-IN",
+        "en-IN",
+        "gu-IN",
+        "as-IN",
+        "ur-IN",
+        "ne-IN",
+        "kok-IN",
+        "ks-IN",
+        "sd-IN",
+        "sa-IN",
+        "sat-IN",
+        "mni-IN",
+        "brx-IN",
+        "mai-IN",
+        "doi-IN",
+    ],
+    typing.Any,
+]