sarvamai 0.1.23a5__py3-none-any.whl → 0.1.23a6__py3-none-any.whl

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/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/requests/speech_to_text_response.py

@@ -1,28 +1,20 @@
 # This file was auto-generated by Fern from our API Definition.
 
 import typing_extensions
-from .diarized_transcript import DiarizedTranscriptParams
-from .timestamps_model import TimestampsModelParams
 
 
 class SpeechToTextResponseParams(typing_extensions.TypedDict):
-    request_id: typing_extensions.NotRequired[str]
-    transcript: str
+    request_id: str
     """
-    The transcribed text from the provided audio file.
+    Unique identifier for the request
     """
 
-    timestamps: typing_extensions.NotRequired[TimestampsModelParams]
-    """
-    Contains timestamps for the transcribed text. This field is included only if with_timestamps is set to true
-    """
-
-    diarized_transcript: typing_extensions.NotRequired[DiarizedTranscriptParams]
+    transcript: str
     """
-    Diarized transcript of the provided speech
+    The transcribed text from the provided audio file.
     """
 
-    language_code: typing_extensions.NotRequired[str]
+    language_code: str
     """
-    This will return the BCP-47 code of language spoken in the input. If multiple languages are detected, this will return language code of most predominant spoken language. If no language is detected, this will be null
+    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/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/requests/speech_to_text_translate_response.py

@@ -1,23 +1,20 @@
 # This file was auto-generated by Fern from our API Definition.
 
 import typing_extensions
-from ..types.speech_to_text_translate_language import SpeechToTextTranslateLanguage
-from .diarized_transcript import DiarizedTranscriptParams
 
 
 class SpeechToTextTranslateResponseParams(typing_extensions.TypedDict):
-    request_id: typing_extensions.NotRequired[str]
-    transcript: str
+    request_id: str
     """
-    Transcript of the provided speech
+    Unique identifier for the request
     """
 
-    language_code: typing_extensions.NotRequired[SpeechToTextTranslateLanguage]
+    transcript: str
     """
-    This will return the BCP-47 code of language spoken in the input. If multiple languages are detected, this will return language code of most predominant spoken language. If no language is detected, this will be null
+    English translation of the provided speech
     """
 
-    diarized_transcript: typing_extensions.NotRequired[DiarizedTranscriptParams]
+    language_code: str
     """
-    Diarized transcript of the provided speech
+    The BCP-47 code of the detected source language spoken in the input (e.g., hi-IN, kn-IN).
     """

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/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/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/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/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/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/speech_to_text_streaming/types/speech_to_text_streaming_language_code.py

@@ -3,6 +3,30 @@
 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"],
+    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/speech_to_text_translate_streaming/__init__.py

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

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/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/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/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/types/speech_to_text_language.py

@@ -4,7 +4,30 @@ 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"
+        "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,
 ]

sarvamai/types/speech_to_text_response.py

@@ -4,30 +4,22 @@ import typing
 
 import pydantic
 from ..core.pydantic_utilities import IS_PYDANTIC_V2, UniversalBaseModel
-from .diarized_transcript import DiarizedTranscript
-from .timestamps_model import TimestampsModel
 
 
 class SpeechToTextResponse(UniversalBaseModel):
-    request_id: typing.Optional[str] = None
-    transcript: str = pydantic.Field()
+    request_id: str = pydantic.Field()
     """
-    The transcribed text from the provided audio file.
+    Unique identifier for the request
     """
 
-    timestamps: typing.Optional[TimestampsModel] = pydantic.Field(default=None)
-    """
-    Contains timestamps for the transcribed text. This field is included only if with_timestamps is set to true
-    """
-
-    diarized_transcript: typing.Optional[DiarizedTranscript] = pydantic.Field(default=None)
+    transcript: str = pydantic.Field()
     """
-    Diarized transcript of the provided speech
+    The transcribed text from the provided audio file.
     """
 
-    language_code: typing.Optional[str] = pydantic.Field(default=None)
+    language_code: str = pydantic.Field()
     """
-    This will return the BCP-47 code of language spoken in the input. If multiple languages are detected, this will return language code of most predominant spoken language. If no language is detected, this will be null
+    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.
     """
 
     if IS_PYDANTIC_V2:

sarvamai/types/speech_to_text_transcription_data.py

@@ -33,6 +33,20 @@ class SpeechToTextTranscriptionData(UniversalBaseModel):
     BCP-47 code of detected language
     """
 
+    language_probability: typing.Optional[float] = pydantic.Field(default=None)
+    """
+    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: TranscriptionMetrics
 
     if IS_PYDANTIC_V2:

sarvamai/types/speech_to_text_translate_language.py

@@ -3,6 +3,30 @@
 import typing
 
 SpeechToTextTranslateLanguage = typing.Union[
-    typing.Literal["hi-IN", "bn-IN", "kn-IN", "ml-IN", "mr-IN", "od-IN", "pa-IN", "ta-IN", "te-IN", "gu-IN", "en-IN"],
+    typing.Literal[
+        "hi-IN",
+        "bn-IN",
+        "kn-IN",
+        "ml-IN",
+        "mr-IN",
+        "od-IN",
+        "pa-IN",
+        "ta-IN",
+        "te-IN",
+        "gu-IN",
+        "en-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/types/speech_to_text_translate_response.py

@@ -4,25 +4,22 @@ import typing
 
 import pydantic
 from ..core.pydantic_utilities import IS_PYDANTIC_V2, UniversalBaseModel
-from .diarized_transcript import DiarizedTranscript
-from .speech_to_text_translate_language import SpeechToTextTranslateLanguage
 
 
 class SpeechToTextTranslateResponse(UniversalBaseModel):
-    request_id: typing.Optional[str] = None
-    transcript: str = pydantic.Field()
+    request_id: str = pydantic.Field()
     """
-    Transcript of the provided speech
+    Unique identifier for the request
     """
 
-    language_code: typing.Optional[SpeechToTextTranslateLanguage] = pydantic.Field(default=None)
+    transcript: str = pydantic.Field()
     """
-    This will return the BCP-47 code of language spoken in the input. If multiple languages are detected, this will return language code of most predominant spoken language. If no language is detected, this will be null
+    English translation of the provided speech
     """
 
-    diarized_transcript: typing.Optional[DiarizedTranscript] = pydantic.Field(default=None)
+    language_code: str = pydantic.Field()
     """
-    Diarized transcript of the provided speech
+    The BCP-47 code of the detected source language spoken in the input (e.g., hi-IN, kn-IN).
     """
 
     if IS_PYDANTIC_V2:

sarvamai/types/speech_to_text_translate_transcription_data.py

@@ -23,6 +23,19 @@ class SpeechToTextTranslateTranscriptionData(UniversalBaseModel):
     BCP-47 code of detected source language (null when language detection is in progress)
     """
 
+    language_probability: typing.Optional[float] = pydantic.Field(default=None)
+    """
+    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: TranscriptionMetrics
 
     if IS_PYDANTIC_V2:

{sarvamai-0.1.23a5.dist-info → sarvamai-0.1.23a6.dist-info}/METADATA

@@ -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.dist-info → sarvamai-0.1.23a6.dist-info}/RECORD

@@ -1,11 +1,11 @@
-sarvamai/__init__.py,sha256=Lm2OWqz5V4YobtV9iX7XnEF7EsXEKg6scXqzdBECti4,11531
+sarvamai/__init__.py,sha256=YgHvuv6dxlTz4zYxWfRcgZKp1imFf0gmwvCQRclWOj0,11721
 sarvamai/chat/__init__.py,sha256=_VhToAyIt_5axN6CLJwtxg3-CO7THa_23pbUzqhXJa4,85
 sarvamai/chat/client.py,sha256=xOSj83Gr6Q7eY2qUeATiuXYQqBqWqSCQlIEopK5fKus,11022
 sarvamai/chat/raw_client.py,sha256=A2kRuZcVWlJhyYCD7YKgqNkZEp3cYa1731KhRkhirU0,17885
 sarvamai/client.py,sha256=J30X_os1lPf8Wml0KDFEf6p8VGHhgF_lf3nw1T2D3qo,8207
 sarvamai/core/__init__.py,sha256=YE2CtXeASe1RAbaI39twKWYKCuT4tW5is9HWHhJjR_g,1653
 sarvamai/core/api_error.py,sha256=44vPoTyWN59gonCIZMdzw7M1uspygiLnr3GNFOoVL2Q,614
-sarvamai/core/client_wrapper.py,sha256=xycjM0SER509jhTOV9GZTxV42go3HQOVrskOdattRwM,2570
+sarvamai/core/client_wrapper.py,sha256=Ijmv6KUlK_OtCy7p9OYCUSgVRv62-xOAyyAvR5uw3AY,2570
 sarvamai/core/datetime_utils.py,sha256=nBys2IsYrhPdszxGKCNRPSOCwa-5DWOHG95FB8G9PKo,1047
 sarvamai/core/events.py,sha256=HvKBdSoYcFetk7cgNXb7FxuY-FtY8NtUhZIN7mGVx8U,1159
 sarvamai/core/file.py,sha256=d4NNbX8XvXP32z8KpK2Xovv33nFfruIrpz0QWxlgpZk,2663
@@ -68,15 +68,15 @@ sarvamai/requests/ping_signal.py,sha256=TSgmfz2k4X1L6TzvX8u2SKZ6XQY3bSf7nPZf8mUV
 sarvamai/requests/send_text.py,sha256=DWzbNgeNN2xSIYgk2zEisgLqjwq5oleqJVHrtOnIqbE,267
 sarvamai/requests/send_text_data.py,sha256=2jds-xd77u-YTgIWQsTUBKE-_7tsrFshXXuC_Ld4ULo,161
 sarvamai/requests/speech_to_text_job_parameters.py,sha256=C1dpp5IkylFXdSNfKDhSo8jbFUquFh3SURFVME9XXq4,2854
-sarvamai/requests/speech_to_text_response.py,sha256=GS3jNmHDOxqNZ7cvftD62khUMSBIQUu6zEPdCqk8zJk,1041
+sarvamai/requests/speech_to_text_response.py,sha256=5Le4BCjYQcf87seNWdrFhHi5YHRJ-ljMLv3nTwsNnWc,517
 sarvamai/requests/speech_to_text_response_data.py,sha256=69fYRdL0tCKpgKQqwzcM4T4Nf_lRxJFh-VCFe_tN964,364
 sarvamai/requests/speech_to_text_streaming_response.py,sha256=cN5tKE9wOWuyBna4wmrf-0LfkOULMpRaJ7qjLuu76V0,348
-sarvamai/requests/speech_to_text_transcription_data.py,sha256=Vc65hXDq65d14cP-fDJm151bi7XEKgPItNGt1UL6cOY,877
+sarvamai/requests/speech_to_text_transcription_data.py,sha256=6YjW2ySX-yIql9MGM02wMue4lNOX1rwif8eSg7jQWo0,1413
 sarvamai/requests/speech_to_text_translate_job_parameters.py,sha256=Q1mL3ul3WYArryp-HP_wHc8WCTLq6ZFhCUFeH3NM80E,1147
-sarvamai/requests/speech_to_text_translate_response.py,sha256=xLV2F37PkGR0erRDfTBEPWvywR8eVSL9JbH5a0C9wkY,893
+sarvamai/requests/speech_to_text_translate_response.py,sha256=pVxKjb9DAtlXM1jXA94TmCAzOrmzIuMjcM-mUwTnDRA,456
 sarvamai/requests/speech_to_text_translate_response_data.py,sha256=OmjunP9R2xertJKn4fmpyzjDdj1_B_Yh6ZjH1eOlR-Q,407
 sarvamai/requests/speech_to_text_translate_streaming_response.py,sha256=KTjYZ0_oLapuM5Iiq7UwejMsrL1TGgFAW4k5l17TkZs,385
-sarvamai/requests/speech_to_text_translate_transcription_data.py,sha256=oAmW5ihTd301IJYN2u2KrZxB0j3EMacFBfvIhtOSjFI,595
+sarvamai/requests/speech_to_text_translate_transcription_data.py,sha256=AEECwgmzR_uInedR38jiksZN8wx_l3sQKQA2dLCi13c,1088
 sarvamai/requests/stop_configuration.py,sha256=Xmp8zyUpnN65pH5A7NqefckB8wk53_BBzOUrgRm2gXs,146
 sarvamai/requests/stt_flush_signal.py,sha256=Gb-SoPPAyVKFVPZKxebLgV4bAv21NjVgvfCl5cqcxrY,360
 sarvamai/requests/task_detail_v_1.py,sha256=2rehl7dSDSgzaw13b9bODamhiN2uB-IK4cOksq8Vmqc,582
@@ -93,14 +93,15 @@ sarvamai/speech_to_text_job/__init__.py,sha256=_VhToAyIt_5axN6CLJwtxg3-CO7THa_23
 sarvamai/speech_to_text_job/client.py,sha256=nMhBJa1rf1qQdQhlB1WUtyzOGhsWxECCPAJwBvWnq8M,18930
 sarvamai/speech_to_text_job/job.py,sha256=9AfVSp5nzrl-Cx_1n2AJZqTMzp6Dkz2cvmbdq78fCgM,18751
 sarvamai/speech_to_text_job/raw_client.py,sha256=6MB82mSqAOi92mE8vUeNSTB0wuxLZYRwizt15R6r-wo,49394
-sarvamai/speech_to_text_streaming/__init__.py,sha256=5l81Q5goyVA8oC5cKaS9-Hv4_PR2nYC318VcmaUTpg8,579
-sarvamai/speech_to_text_streaming/client.py,sha256=KiJEcJuHYc7bfbOXIBOEras0KjM2zhgVYo4lUK4Hlek,14801
-sarvamai/speech_to_text_streaming/raw_client.py,sha256=QzC3ytldXbH6b5-tpPT7tmie5VhaXSGd0J1RbvspnaY,13982
+sarvamai/speech_to_text_streaming/__init__.py,sha256=D_WTGMhL_12vOb4IazZpC3o91HKFgPRNjs2r0EEyPBk,665
+sarvamai/speech_to_text_streaming/client.py,sha256=-6fmOpPZXu6iRHSrWCU6mrWUhE24_AIzyBj9yljpxfY,16686
+sarvamai/speech_to_text_streaming/raw_client.py,sha256=jJIiKxj4QmU3HAu-ZVk7vZnPWO3kOXXUXVWT37o6hIg,15867
 sarvamai/speech_to_text_streaming/socket_client.py,sha256=P6qXRN0s3UFAp6CP5lkqrW2KPK9me70ZVfWquxLB4wI,7538
-sarvamai/speech_to_text_streaming/types/__init__.py,sha256=M6FNnzx7E98J7AzP2oU_94qVsWoxaYZG9_Syc5p5qQg,825
+sarvamai/speech_to_text_streaming/types/__init__.py,sha256=u6sdJI-GFD-CAKX7xFOeS64qwbDW5xuICovpYwBzwLY,962
 sarvamai/speech_to_text_streaming/types/speech_to_text_streaming_flush_signal.py,sha256=dDJOBlzAjhuiSVqW2RHHY1f6xy0DU_Yoo9UV8-7MjnA,173
 sarvamai/speech_to_text_streaming/types/speech_to_text_streaming_high_vad_sensitivity.py,sha256=OwPwffa8TkLPGMnOTn5S7d-HmV8QmN3B7fHz8I1-VT8,180
-sarvamai/speech_to_text_streaming/types/speech_to_text_streaming_language_code.py,sha256=LxgEifmgWTCFZn9U-f-TWKxRPng3a2J26Zt526QrA0Y,267
+sarvamai/speech_to_text_streaming/types/speech_to_text_streaming_input_audio_codec.py,sha256=dxnhjo9zo9WFk8CVUklubPhMaTxlaRvgFtuMmeQcAqo,208
+sarvamai/speech_to_text_streaming/types/speech_to_text_streaming_language_code.py,sha256=7JdW_xi7IljKpdpOfwgNqQnfpng8VwASVPJ4QuOMh24,572
 sarvamai/speech_to_text_streaming/types/speech_to_text_streaming_mode.py,sha256=jce75h2M9dEMD_eC29zCLQZFbLfCy8sdxEIISxtfubQ,217
 sarvamai/speech_to_text_streaming/types/speech_to_text_streaming_model.py,sha256=CpWC1HmcPJKWCSHhTYSaC8_pMsfNTBA-EHq-sfCjS-A,179
 sarvamai/speech_to_text_streaming/types/speech_to_text_streaming_vad_signals.py,sha256=8wiFOB7WDMbYCcMTYgNFJaIjEytYeXpJLwr_O_mH0TI,172
@@ -108,13 +109,14 @@ sarvamai/speech_to_text_translate_job/__init__.py,sha256=_VhToAyIt_5axN6CLJwtxg3
 sarvamai/speech_to_text_translate_job/client.py,sha256=xu8kYtCESDB7LzL8YKBUq5qhTPMIl3_H3XD2L_7y4UU,18969
 sarvamai/speech_to_text_translate_job/job.py,sha256=tL1Zemsogb_AK9wqZwN4ooPaN176sFKduTH9g87y-WU,18938
 sarvamai/speech_to_text_translate_job/raw_client.py,sha256=Emx14cRiAZXg1PqZkoJbDOKwyDmOgwxWlqPkAPZ9GPU,50797
-sarvamai/speech_to_text_translate_streaming/__init__.py,sha256=HYq3MzUyWa4Kt1ou6vgAkbMCHoIcXzoqn6V1kya5c6g,423
-sarvamai/speech_to_text_translate_streaming/client.py,sha256=wFAn-mrcOUFFa5sAgjSkAx9OZiTNbQvyJwBlid3hJP8,10757
-sarvamai/speech_to_text_translate_streaming/raw_client.py,sha256=jDUuPdYWeh-sxeNHdEzLeMaAyCKJZDz-5mrVRUPoFZU,9866
+sarvamai/speech_to_text_translate_streaming/__init__.py,sha256=s6HPwrkABpkhDSsd_t6pVRiWfY4MfVE0lVj9b4V_fx4,527
+sarvamai/speech_to_text_translate_streaming/client.py,sha256=QiKQWWjX4iVTLF6YwmeJKcqzU9TNFzENCJeI4xk2ndQ,11813
+sarvamai/speech_to_text_translate_streaming/raw_client.py,sha256=lUiORV21loTC3Fq5m8D_XJhxGdFFOfnmDNem3v2B64Y,10922
 sarvamai/speech_to_text_translate_streaming/socket_client.py,sha256=ipEPSj5eHAyDpuEXfaP7JJL1rXJXGEo-IB888ReAFKs,8901
-sarvamai/speech_to_text_translate_streaming/types/__init__.py,sha256=NFfpVz1gILXj96rWMFw4MgSpni0Yvs8CHgwJ9Xry7OU,575
+sarvamai/speech_to_text_translate_streaming/types/__init__.py,sha256=nsKmvwkhcPekF9kcStDhTDilALFf2jT-wfCn25KVe7U,740
 sarvamai/speech_to_text_translate_streaming/types/speech_to_text_translate_streaming_flush_signal.py,sha256=jkjvCGJ1pFKi3AOTkwMW-lo18WGgrgAhMpoe5P0AMzA,182
 sarvamai/speech_to_text_translate_streaming/types/speech_to_text_translate_streaming_high_vad_sensitivity.py,sha256=r6MvTlkM0VEpb4dpnMHtINOZ-gYc22o0Fx_Xce2rjvo,189
+sarvamai/speech_to_text_translate_streaming/types/speech_to_text_translate_streaming_input_audio_codec.py,sha256=x8K3YSZ_mVkfiyhmRk8RvEInw3KTOwCapj5q2tOPggo,217
 sarvamai/speech_to_text_translate_streaming/types/speech_to_text_translate_streaming_vad_signals.py,sha256=EV3xd9qyKMnMvA9rO-qFDDIac4b84roBu7n-maaPxG8,181
 sarvamai/text/__init__.py,sha256=_VhToAyIt_5axN6CLJwtxg3-CO7THa_23pbUzqhXJa4,85
 sarvamai/text/client.py,sha256=jNOHjc3Pu7xsnuDMuf7xX3WHAIO7B-AgMgKj3pL3eZA,29657
@@ -188,19 +190,19 @@ sarvamai/types/send_text.py,sha256=kgST6V5NuURzgBpuiDi8fVwDg768ttDoeY9k1dSSb1Y,6
 sarvamai/types/send_text_data.py,sha256=H8yfcvd4gvyN34RrZ9i4qQvieednNBhL7i7isX4asuY,519
 sarvamai/types/speech_sample_rate.py,sha256=Hfi79KL2Y1W7OIvvrfWnt7EUvmU5i7bxYvXivrY_aUA,88
 sarvamai/types/speech_to_text_job_parameters.py,sha256=JFKO2rIyBJE_PK4mrMMwY4e562l_6HbZtP2kd4CbKa8,3293
-sarvamai/types/speech_to_text_language.py,sha256=cq8FBOX0DfYB3v8jgNteQtHeJcqWqzKWJVyYGwwo_w0,279
+sarvamai/types/speech_to_text_language.py,sha256=dYmAo9lg7iC0w_GVjmnYmALTQgJjma2wFDXQY3_Q4pc,578
 sarvamai/types/speech_to_text_model.py,sha256=hHC3aOXzdPt8i32qJw4ZLz5bdREWVQl4P7Y_lOURJu4,170
-sarvamai/types/speech_to_text_response.py,sha256=iWRGEJeHUFIOxeEhoCQu68njeA6lcqXbT2czV-O8Wx0,1438
+sarvamai/types/speech_to_text_response.py,sha256=0tXgHr-1zN54hA_6nUkMV5Jkt7pEmjl7zJQ8gOt5Nq8,932
 sarvamai/types/speech_to_text_response_data.py,sha256=gbxZTBSjbN3ZIa10K6tWPYtymcpnQTFIaUnXkOmsmD4,322
 sarvamai/types/speech_to_text_streaming_response.py,sha256=z6tVAHbVK9lC3w3lac__LEUfO8AAzEilkeGlaLskTtc,687
-sarvamai/types/speech_to_text_transcription_data.py,sha256=EqwPAPSi98PwARaTj-ufzFUSHyN-NPoPla5vi_KERrU,1297
+sarvamai/types/speech_to_text_transcription_data.py,sha256=nrGmNnA9Au5CpBTIyG9wto2PSl2I7T8Kll4KWOkCrFg,1850
 sarvamai/types/speech_to_text_translate_job_parameters.py,sha256=-E85BoIBxW5Ck638aRFE0fC_f43RCoIkboAFu2QlBBs,1566
-sarvamai/types/speech_to_text_translate_language.py,sha256=yikNM-roIumVG-eqBWss93wLGudZdLPwd0i3VcXH5zo,263
+sarvamai/types/speech_to_text_translate_language.py,sha256=lmWyAWMwSSDNPU5HrJtmhNYLRPhWo41ShMUCrXyEPoc,568
 sarvamai/types/speech_to_text_translate_model.py,sha256=CVSz6gJBY82GhhEuWSdzRLJW9XTsAgweRnKd1tN6mXo,139
-sarvamai/types/speech_to_text_translate_response.py,sha256=Z5Na7IQW2ok3TP21xd-jKkwioplEKfonNIMhoJQKkVw,1278
+sarvamai/types/speech_to_text_translate_response.py,sha256=v1xTwIva81c74hWy9-ipbJFQWvXlAvw_b7o9xvo6mgc,871
 sarvamai/types/speech_to_text_translate_response_data.py,sha256=_NlLVp7oQU3em_4E47QVbIP9nromPE07Z9HtMpY1lrU,359
 sarvamai/types/speech_to_text_translate_streaming_response.py,sha256=J6h3AGdAJxpODFs30bR-e6OaWKa__oVhwv_TrbPSO98,724
-sarvamai/types/speech_to_text_translate_transcription_data.py,sha256=-cZZm21um6erOzx18nAgBYKA3Qc3YzNADUCnfx9mD-k,996
+sarvamai/types/speech_to_text_translate_transcription_data.py,sha256=HR-7y3LKVt4e3FY028aEXQmORtjdpjBY29AWTIRsRVA,1506
 sarvamai/types/spoken_form_numerals_format.py,sha256=soBly93wMkazIcp2GDM0Mf1MjY140Pe24hBlwNoWge0,169
 sarvamai/types/stop_configuration.py,sha256=yA_q4s4BIrbl3FotZpg4ZcyL10C7gVI0s2dqvH32BNw,136
 sarvamai/types/storage_container_type.py,sha256=DZXDiDj74lMmUq6jaZfIMW1zMXgoVdY6rs_FcyB9OGk,184
@@ -226,6 +228,6 @@ sarvamai/types/transliterate_mode.py,sha256=1jSEMlGcoLkWuk12TgoOpSgwifa4rThGKZ1h
 sarvamai/types/transliterate_source_language.py,sha256=bSY9wJszF0sg-Cgg6F-YcWC8ly1mIlj9rqa15-jBtx8,283
 sarvamai/types/transliteration_response.py,sha256=yt-lzTbDeJ_ZL4I8kQa6oESxA9ebeJJY7LfFHpdEsmM,815
 sarvamai/version.py,sha256=Qkp3Ee9YH-O9RTix90e0i7iNrFAGN-QDt2AFwGA4n8k,75
-sarvamai-0.1.23a5.dist-info/METADATA,sha256=nPZ4wzgXrEkNFWPm-2mlOxCW4n4xIFhbBehaDWxkTiE,26753
-sarvamai-0.1.23a5.dist-info/WHEEL,sha256=Zb28QaM1gQi8f4VCBhsUklF61CTlNYfs9YAZn-TOGFk,88
-sarvamai-0.1.23a5.dist-info/RECORD,,
+sarvamai-0.1.23a6.dist-info/METADATA,sha256=5pdMsPKXwU2rQqBN-U7GmnWbAKVgSLmtfd7E0G8e6KU,26753
+sarvamai-0.1.23a6.dist-info/WHEEL,sha256=Zb28QaM1gQi8f4VCBhsUklF61CTlNYfs9YAZn-TOGFk,88
+sarvamai-0.1.23a6.dist-info/RECORD,,