sarvamai 0.1.20a3__py3-none-any.whl → 0.1.22a1__py3-none-any.whl

sarvamai/__init__.py

@@ -56,6 +56,7 @@ from .types import (
     NumeralsFormat,
     PingSignal,
     ReasoningEffort,
+    ResponseSpeechState,
     ResponseType,
     Role,
     SarvamModelIds,
@@ -297,6 +298,7 @@ __all__ = [
     "PingSignal",
     "PingSignalParams",
     "ReasoningEffort",
+    "ResponseSpeechState",
     "ResponseType",
     "Role",
     "SarvamAI",

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/0.1.20a3",
+            "User-Agent": "sarvamai/0.1.22a1",
             "X-Fern-Language": "Python",
             "X-Fern-SDK-Name": "sarvamai",
-            "X-Fern-SDK-Version": "0.1.20a3",
+            "X-Fern-SDK-Version": "0.1.22a1",
             **(self.get_custom_headers() or {}),
         }
         headers["api-subscription-key"] = self.api_subscription_key

sarvamai/requests/speech_to_text_transcription_data.py

@@ -3,6 +3,7 @@
 import typing
 
 import typing_extensions
+from ..types.response_speech_state import ResponseSpeechState
 from .transcription_metrics import TranscriptionMetricsParams
 
 
@@ -32,4 +33,9 @@ class SpeechToTextTranscriptionDataParams(typing_extensions.TypedDict):
     BCP-47 code of detected language
     """
 
+    response_speech_state: typing_extensions.NotRequired[ResponseSpeechState]
+    """
+    Current state of speech detection and processing
+    """
+
     metrics: TranscriptionMetricsParams

sarvamai/requests/speech_to_text_translate_transcription_data.py

@@ -1,6 +1,7 @@
 # This file was auto-generated by Fern from our API Definition.
 
 import typing_extensions
+from ..types.response_speech_state import ResponseSpeechState
 from .transcription_metrics import TranscriptionMetricsParams
 
 
@@ -20,4 +21,9 @@ class SpeechToTextTranslateTranscriptionDataParams(typing_extensions.TypedDict):
     BCP-47 code of detected source language (null when language detection is in progress)
     """
 
+    response_speech_state: typing_extensions.NotRequired[ResponseSpeechState]
+    """
+    Current state of speech detection and processing
+    """
+
     metrics: TranscriptionMetricsParams

sarvamai/speech_to_text_streaming/client.py

@@ -49,11 +49,17 @@ class SpeechToTextStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextStreamingFlushSignal] = None,
+        stream_ongoing_speech_results: typing.Optional[str] = None,
+        streaming_ongoing_requests_frame_size: typing.Optional[str] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[SpeechToTextStreamingSocketClient]:
         """
-        WebSocket channel for real-time speech to text streaming
+        WebSocket channel for real-time speech to text streaming.
+
+        **Note:** This API Reference page is provided for informational purposes only.
+        The Try It playground may not provide the best experience for streaming audio.
+        For optimal streaming performance, please use the SDK or implement your own WebSocket client.
 
         Parameters
         ----------
@@ -78,6 +84,12 @@ class SpeechToTextStreamingClient:
         flush_signal : typing.Optional[SpeechToTextStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription
 
+        stream_ongoing_speech_results : typing.Optional[str]
+            Enable streaming of ongoing speech results during active speech
+
+        streaming_ongoing_requests_frame_size : typing.Optional[str]
+            Frame size for streaming ongoing speech results (1-100)
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -104,6 +116,12 @@ 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 stream_ongoing_speech_results is not None:
+            query_params = query_params.add("stream_ongoing_speech_results", stream_ongoing_speech_results)
+        if streaming_ongoing_requests_frame_size is not None:
+            query_params = query_params.add(
+                "streaming_ongoing_requests_frame_size", streaming_ongoing_requests_frame_size
+            )
         ws_url = ws_url + f"?{query_params}"
         headers = self._raw_client._client_wrapper.get_headers()
         if api_subscription_key is not None:
@@ -154,11 +172,17 @@ class AsyncSpeechToTextStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextStreamingFlushSignal] = None,
+        stream_ongoing_speech_results: typing.Optional[str] = None,
+        streaming_ongoing_requests_frame_size: typing.Optional[str] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncSpeechToTextStreamingSocketClient]:
         """
-        WebSocket channel for real-time speech to text streaming
+        WebSocket channel for real-time speech to text streaming.
+
+        **Note:** This API Reference page is provided for informational purposes only.
+        The Try It playground may not provide the best experience for streaming audio.
+        For optimal streaming performance, please use the SDK or implement your own WebSocket client.
 
         Parameters
         ----------
@@ -183,6 +207,12 @@ class AsyncSpeechToTextStreamingClient:
         flush_signal : typing.Optional[SpeechToTextStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription
 
+        stream_ongoing_speech_results : typing.Optional[str]
+            Enable streaming of ongoing speech results during active speech
+
+        streaming_ongoing_requests_frame_size : typing.Optional[str]
+            Frame size for streaming ongoing speech results (1-100)
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -209,6 +239,12 @@ 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 stream_ongoing_speech_results is not None:
+            query_params = query_params.add("stream_ongoing_speech_results", stream_ongoing_speech_results)
+        if streaming_ongoing_requests_frame_size is not None:
+            query_params = query_params.add(
+                "streaming_ongoing_requests_frame_size", streaming_ongoing_requests_frame_size
+            )
         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

@@ -37,11 +37,17 @@ class RawSpeechToTextStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextStreamingFlushSignal] = None,
+        stream_ongoing_speech_results: typing.Optional[str] = None,
+        streaming_ongoing_requests_frame_size: typing.Optional[str] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[SpeechToTextStreamingSocketClient]:
         """
-        WebSocket channel for real-time speech to text streaming
+        WebSocket channel for real-time speech to text streaming.
+
+        **Note:** This API Reference page is provided for informational purposes only.
+        The Try It playground may not provide the best experience for streaming audio.
+        For optimal streaming performance, please use the SDK or implement your own WebSocket client.
 
         Parameters
         ----------
@@ -66,6 +72,12 @@ class RawSpeechToTextStreamingClient:
         flush_signal : typing.Optional[SpeechToTextStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription
 
+        stream_ongoing_speech_results : typing.Optional[str]
+            Enable streaming of ongoing speech results during active speech
+
+        streaming_ongoing_requests_frame_size : typing.Optional[str]
+            Frame size for streaming ongoing speech results (1-100)
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -92,6 +104,12 @@ 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 stream_ongoing_speech_results is not None:
+            query_params = query_params.add("stream_ongoing_speech_results", stream_ongoing_speech_results)
+        if streaming_ongoing_requests_frame_size is not None:
+            query_params = query_params.add(
+                "streaming_ongoing_requests_frame_size", streaming_ongoing_requests_frame_size
+            )
         ws_url = ws_url + f"?{query_params}"
         headers = self._client_wrapper.get_headers()
         if api_subscription_key is not None:
@@ -131,11 +149,17 @@ class AsyncRawSpeechToTextStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextStreamingFlushSignal] = None,
+        stream_ongoing_speech_results: typing.Optional[str] = None,
+        streaming_ongoing_requests_frame_size: typing.Optional[str] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncSpeechToTextStreamingSocketClient]:
         """
-        WebSocket channel for real-time speech to text streaming
+        WebSocket channel for real-time speech to text streaming.
+
+        **Note:** This API Reference page is provided for informational purposes only.
+        The Try It playground may not provide the best experience for streaming audio.
+        For optimal streaming performance, please use the SDK or implement your own WebSocket client.
 
         Parameters
         ----------
@@ -160,6 +184,12 @@ class AsyncRawSpeechToTextStreamingClient:
         flush_signal : typing.Optional[SpeechToTextStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription
 
+        stream_ongoing_speech_results : typing.Optional[str]
+            Enable streaming of ongoing speech results during active speech
+
+        streaming_ongoing_requests_frame_size : typing.Optional[str]
+            Frame size for streaming ongoing speech results (1-100)
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -186,6 +216,12 @@ 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 stream_ongoing_speech_results is not None:
+            query_params = query_params.add("stream_ongoing_speech_results", stream_ongoing_speech_results)
+        if streaming_ongoing_requests_frame_size is not None:
+            query_params = query_params.add(
+                "streaming_ongoing_requests_frame_size", streaming_ongoing_requests_frame_size
+            )
         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/client.py

@@ -49,11 +49,17 @@ class SpeechToTextTranslateStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextTranslateStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextTranslateStreamingFlushSignal] = None,
+        stream_ongoing_speech_results: typing.Optional[str] = None,
+        streaming_ongoing_requests_frame_size: typing.Optional[str] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[SpeechToTextTranslateStreamingSocketClient]:
         """
-        WebSocket channel for real-time speech to text streaming with English translation
+        WebSocket channel for real-time speech to text streaming with English translation.
+
+        **Note:** This API Reference page is provided for informational purposes only.
+        The Try It playground may not provide the best experience for streaming audio.
+        For optimal streaming performance, please use the SDK or implement your own WebSocket client.
 
         Parameters
         ----------
@@ -75,6 +81,12 @@ class SpeechToTextTranslateStreamingClient:
         flush_signal : typing.Optional[SpeechToTextTranslateStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription and translation
 
+        stream_ongoing_speech_results : typing.Optional[str]
+            Enable streaming of ongoing speech results during active speech
+
+        streaming_ongoing_requests_frame_size : typing.Optional[str]
+            Frame size for streaming ongoing speech results (1-100)
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -99,6 +111,12 @@ 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 stream_ongoing_speech_results is not None:
+            query_params = query_params.add("stream_ongoing_speech_results", stream_ongoing_speech_results)
+        if streaming_ongoing_requests_frame_size is not None:
+            query_params = query_params.add(
+                "streaming_ongoing_requests_frame_size", streaming_ongoing_requests_frame_size
+            )
         ws_url = ws_url + f"?{query_params}"
         headers = self._raw_client._client_wrapper.get_headers()
         if api_subscription_key is not None:
@@ -148,11 +166,17 @@ class AsyncSpeechToTextTranslateStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextTranslateStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextTranslateStreamingFlushSignal] = None,
+        stream_ongoing_speech_results: typing.Optional[str] = None,
+        streaming_ongoing_requests_frame_size: typing.Optional[str] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncSpeechToTextTranslateStreamingSocketClient]:
         """
-        WebSocket channel for real-time speech to text streaming with English translation
+        WebSocket channel for real-time speech to text streaming with English translation.
+
+        **Note:** This API Reference page is provided for informational purposes only.
+        The Try It playground may not provide the best experience for streaming audio.
+        For optimal streaming performance, please use the SDK or implement your own WebSocket client.
 
         Parameters
         ----------
@@ -174,6 +198,12 @@ class AsyncSpeechToTextTranslateStreamingClient:
         flush_signal : typing.Optional[SpeechToTextTranslateStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription and translation
 
+        stream_ongoing_speech_results : typing.Optional[str]
+            Enable streaming of ongoing speech results during active speech
+
+        streaming_ongoing_requests_frame_size : typing.Optional[str]
+            Frame size for streaming ongoing speech results (1-100)
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -198,6 +228,12 @@ 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 stream_ongoing_speech_results is not None:
+            query_params = query_params.add("stream_ongoing_speech_results", stream_ongoing_speech_results)
+        if streaming_ongoing_requests_frame_size is not None:
+            query_params = query_params.add(
+                "streaming_ongoing_requests_frame_size", streaming_ongoing_requests_frame_size
+            )
         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

@@ -37,11 +37,17 @@ class RawSpeechToTextTranslateStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextTranslateStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextTranslateStreamingFlushSignal] = None,
+        stream_ongoing_speech_results: typing.Optional[str] = None,
+        streaming_ongoing_requests_frame_size: typing.Optional[str] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[SpeechToTextTranslateStreamingSocketClient]:
         """
-        WebSocket channel for real-time speech to text streaming with English translation
+        WebSocket channel for real-time speech to text streaming with English translation.
+
+        **Note:** This API Reference page is provided for informational purposes only.
+        The Try It playground may not provide the best experience for streaming audio.
+        For optimal streaming performance, please use the SDK or implement your own WebSocket client.
 
         Parameters
         ----------
@@ -63,6 +69,12 @@ class RawSpeechToTextTranslateStreamingClient:
         flush_signal : typing.Optional[SpeechToTextTranslateStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription and translation
 
+        stream_ongoing_speech_results : typing.Optional[str]
+            Enable streaming of ongoing speech results during active speech
+
+        streaming_ongoing_requests_frame_size : typing.Optional[str]
+            Frame size for streaming ongoing speech results (1-100)
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -87,6 +99,12 @@ 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 stream_ongoing_speech_results is not None:
+            query_params = query_params.add("stream_ongoing_speech_results", stream_ongoing_speech_results)
+        if streaming_ongoing_requests_frame_size is not None:
+            query_params = query_params.add(
+                "streaming_ongoing_requests_frame_size", streaming_ongoing_requests_frame_size
+            )
         ws_url = ws_url + f"?{query_params}"
         headers = self._client_wrapper.get_headers()
         if api_subscription_key is not None:
@@ -125,11 +143,17 @@ class AsyncRawSpeechToTextTranslateStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextTranslateStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextTranslateStreamingFlushSignal] = None,
+        stream_ongoing_speech_results: typing.Optional[str] = None,
+        streaming_ongoing_requests_frame_size: typing.Optional[str] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncSpeechToTextTranslateStreamingSocketClient]:
         """
-        WebSocket channel for real-time speech to text streaming with English translation
+        WebSocket channel for real-time speech to text streaming with English translation.
+
+        **Note:** This API Reference page is provided for informational purposes only.
+        The Try It playground may not provide the best experience for streaming audio.
+        For optimal streaming performance, please use the SDK or implement your own WebSocket client.
 
         Parameters
         ----------
@@ -151,6 +175,12 @@ class AsyncRawSpeechToTextTranslateStreamingClient:
         flush_signal : typing.Optional[SpeechToTextTranslateStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription and translation
 
+        stream_ongoing_speech_results : typing.Optional[str]
+            Enable streaming of ongoing speech results during active speech
+
+        streaming_ongoing_requests_frame_size : typing.Optional[str]
+            Frame size for streaming ongoing speech results (1-100)
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -175,6 +205,12 @@ 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 stream_ongoing_speech_results is not None:
+            query_params = query_params.add("stream_ongoing_speech_results", stream_ongoing_speech_results)
+        if streaming_ongoing_requests_frame_size is not None:
+            query_params = query_params.add(
+                "streaming_ongoing_requests_frame_size", streaming_ongoing_requests_frame_size
+            )
         ws_url = ws_url + f"?{query_params}"
         headers = self._client_wrapper.get_headers()
         if api_subscription_key is not None:

sarvamai/text_to_speech_streaming/client.py

@@ -44,8 +44,11 @@ class TextToSpeechStreamingClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[TextToSpeechStreamingSocketClient]:
         """
-        Bidirectional WebSocket channel for real-time TTS synthesis.
-        Supports streaming, flushing, config updates, and audio playback.
+        WebSocket channel for real-time TTS synthesis.
+
+        **Note:** This API Reference page is provided for informational purposes only.
+        The Try It playground may not provide the best experience for streaming audio.
+        For optimal streaming performance, please use the SDK or implement your own WebSocket client.
 
         Parameters
         ----------
@@ -120,8 +123,11 @@ class AsyncTextToSpeechStreamingClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncTextToSpeechStreamingSocketClient]:
         """
-        Bidirectional WebSocket channel for real-time TTS synthesis.
-        Supports streaming, flushing, config updates, and audio playback.
+        WebSocket channel for real-time TTS synthesis.
+
+        **Note:** This API Reference page is provided for informational purposes only.
+        The Try It playground may not provide the best experience for streaming audio.
+        For optimal streaming performance, please use the SDK or implement your own WebSocket client.
 
         Parameters
         ----------

sarvamai/text_to_speech_streaming/raw_client.py

@@ -32,8 +32,11 @@ class RawTextToSpeechStreamingClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[TextToSpeechStreamingSocketClient]:
         """
-        Bidirectional WebSocket channel for real-time TTS synthesis.
-        Supports streaming, flushing, config updates, and audio playback.
+        WebSocket channel for real-time TTS synthesis.
+
+        **Note:** This API Reference page is provided for informational purposes only.
+        The Try It playground may not provide the best experience for streaming audio.
+        For optimal streaming performance, please use the SDK or implement your own WebSocket client.
 
         Parameters
         ----------
@@ -97,8 +100,11 @@ class AsyncRawTextToSpeechStreamingClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncTextToSpeechStreamingSocketClient]:
         """
-        Bidirectional WebSocket channel for real-time TTS synthesis.
-        Supports streaming, flushing, config updates, and audio playback.
+        WebSocket channel for real-time TTS synthesis.
+
+        **Note:** This API Reference page is provided for informational purposes only.
+        The Try It playground may not provide the best experience for streaming audio.
+        For optimal streaming performance, please use the SDK or implement your own WebSocket client.
 
         Parameters
         ----------

sarvamai/types/__init__.py

@@ -57,6 +57,7 @@ from .language_identification_response import LanguageIdentificationResponse
 from .numerals_format import NumeralsFormat
 from .ping_signal import PingSignal
 from .reasoning_effort import ReasoningEffort
+from .response_speech_state import ResponseSpeechState
 from .response_type import ResponseType
 from .role import Role
 from .sarvam_model_ids import SarvamModelIds
@@ -156,6 +157,7 @@ __all__ = [
     "NumeralsFormat",
     "PingSignal",
     "ReasoningEffort",
+    "ResponseSpeechState",
     "ResponseType",
     "Role",
     "SarvamModelIds",

sarvamai/types/completion_event_flag.py

@@ -1,5 +1,3 @@
 # This file was auto-generated by Fern from our API Definition.
 
-import typing
-
-CompletionEventFlag = typing.Union[typing.Literal["true", "false"], typing.Any]
+CompletionEventFlag = bool

sarvamai/types/response_speech_state.py

@@ -0,0 +1,7 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+ResponseSpeechState = typing.Union[
+    typing.Literal["START_SPEECH_TENTATIVE", "VAD_MISFIRE", "START_SPEECH", "END_SPEECH", "ONGOING"], typing.Any
+]

sarvamai/types/speech_to_text_transcription_data.py

@@ -4,6 +4,7 @@ import typing
 
 import pydantic
 from ..core.pydantic_utilities import IS_PYDANTIC_V2, UniversalBaseModel
+from .response_speech_state import ResponseSpeechState
 from .transcription_metrics import TranscriptionMetrics
 
 
@@ -33,6 +34,11 @@ class SpeechToTextTranscriptionData(UniversalBaseModel):
     BCP-47 code of detected language
     """
 
+    response_speech_state: typing.Optional[ResponseSpeechState] = pydantic.Field(default=None)
+    """
+    Current state of speech detection and processing
+    """
+
     metrics: TranscriptionMetrics
 
     if IS_PYDANTIC_V2:

sarvamai/types/speech_to_text_translate_transcription_data.py

@@ -4,6 +4,7 @@ import typing
 
 import pydantic
 from ..core.pydantic_utilities import IS_PYDANTIC_V2, UniversalBaseModel
+from .response_speech_state import ResponseSpeechState
 from .transcription_metrics import TranscriptionMetrics
 
 
@@ -23,6 +24,11 @@ class SpeechToTextTranslateTranscriptionData(UniversalBaseModel):
     BCP-47 code of detected source language (null when language detection is in progress)
     """
 
+    response_speech_state: typing.Optional[ResponseSpeechState] = pydantic.Field(default=None)
+    """
+    Current state of speech detection and processing
+    """
+
     metrics: TranscriptionMetrics
 
     if IS_PYDANTIC_V2:

{sarvamai-0.1.20a3.dist-info → sarvamai-0.1.22a1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sarvamai
-Version: 0.1.20a3
+Version: 0.1.22a1
 Summary: 
 Requires-Python: >=3.8,<4.0
 Classifier: Intended Audience :: Developers

{sarvamai-0.1.20a3.dist-info → sarvamai-0.1.22a1.dist-info}/RECORD

@@ -1,11 +1,11 @@
-sarvamai/__init__.py,sha256=vI_MiA00kkNnsVNJmIFkf1o_-vLIp8SeNy4xdzjQ_wY,11499
+sarvamai/__init__.py,sha256=IQQBebJtLIgp9Nkbs8oLDb68dQoO-fg9Er6yN__-eVw,11551
 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=ZSinVWwLcEoiBinSQsL_LCNa37D6WzIjGqnNN450h8I,2570
+sarvamai/core/client_wrapper.py,sha256=p6m2l7EbyUKsHP83xGTq1y3npx-n_VdF3sD_7NpBd84,2570
 sarvamai/core/datetime_utils.py,sha256=nBys2IsYrhPdszxGKCNRPSOCwa-5DWOHG95FB8G9PKo,1047
 sarvamai/core/events.py,sha256=HvKBdSoYcFetk7cgNXb7FxuY-FtY8NtUhZIN7mGVx8U,1159
 sarvamai/core/file.py,sha256=d4NNbX8XvXP32z8KpK2Xovv33nFfruIrpz0QWxlgpZk,2663
@@ -71,12 +71,12 @@ sarvamai/requests/speech_to_text_job_parameters.py,sha256=F3V6XvdUQlWlB4-5XAcRsF
 sarvamai/requests/speech_to_text_response.py,sha256=GS3jNmHDOxqNZ7cvftD62khUMSBIQUu6zEPdCqk8zJk,1041
 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=ZbrUy4SBEx8OoeOW1PwTbTq8171Z2z2H1Mh_ZZxrREI,1087
 sarvamai/requests/speech_to_text_translate_job_parameters.py,sha256=Cco38i8IhX00S2eW3MhLekqUFMS7hZW2AwbpWyCAgpU,990
 sarvamai/requests/speech_to_text_translate_response.py,sha256=xLV2F37PkGR0erRDfTBEPWvywR8eVSL9JbH5a0C9wkY,893
 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=xBVSuFtYz-tNiE69OMFpfBnxbEG3ZMl7YMFDWoJyhWU,805
 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
@@ -94,8 +94,8 @@ sarvamai/speech_to_text_job/client.py,sha256=WSGBJxYcNxl77Zd1X6VVWjg4zshqecXf6WC
 sarvamai/speech_to_text_job/job.py,sha256=K8HOmwrYd6l82-MZfWDBmNkZeeERyg9YOihnFfvl-Js,15021
 sarvamai/speech_to_text_job/raw_client.py,sha256=OZTPzMhAn-ckE_xKzfZ9QLsEX5EZVOJS0Pf-PBa19jM,48200
 sarvamai/speech_to_text_streaming/__init__.py,sha256=-7nN6AJFryjSvGHVbajYEt-vni6kNDfJUiZJFNl_ao4,535
-sarvamai/speech_to_text_streaming/client.py,sha256=s-0rKrNN5Phk7xR2WOGNoG6zSA00dd_3zBXEfVFdNqk,11176
-sarvamai/speech_to_text_streaming/raw_client.py,sha256=Oe9dz6hBTRkdfIDUGHuDhuCqJooCGdM_hxc9Z2DFfqs,10357
+sarvamai/speech_to_text_streaming/client.py,sha256=z1iJo_9Fc0Fp1c-n4TCN_StmsSPMZUUxHPbGHWwoekg,13328
+sarvamai/speech_to_text_streaming/raw_client.py,sha256=Koy0Repnt3Niws3CrUZAdIRgGcjAma0A08h3TTBBFVs,12509
 sarvamai/speech_to_text_streaming/socket_client.py,sha256=P6qXRN0s3UFAp6CP5lkqrW2KPK9me70ZVfWquxLB4wI,7538
 sarvamai/speech_to_text_streaming/types/__init__.py,sha256=_G5TSTthsnjGmwdV4fpsybjEWMMTNkh-kWXZjgK5X48,755
 sarvamai/speech_to_text_streaming/types/speech_to_text_streaming_flush_signal.py,sha256=dDJOBlzAjhuiSVqW2RHHY1f6xy0DU_Yoo9UV8-7MjnA,173
@@ -108,8 +108,8 @@ sarvamai/speech_to_text_translate_job/client.py,sha256=xu8kYtCESDB7LzL8YKBUq5qhT
 sarvamai/speech_to_text_translate_job/job.py,sha256=DU4k3eB28V8N16M_QEchakVng4IOul6_Qrdn3FumgHA,15208
 sarvamai/speech_to_text_translate_job/raw_client.py,sha256=dAitbu2B9afPK6iT9zNjUJnE5BIr5-lrAlwrfwFxdkU,49507
 sarvamai/speech_to_text_translate_streaming/__init__.py,sha256=s6HPwrkABpkhDSsd_t6pVRiWfY4MfVE0lVj9b4V_fx4,527
-sarvamai/speech_to_text_translate_streaming/client.py,sha256=p8iZeb5cM8QfHIp_1Z36CLJLnkgn81wKCItLdy0-Mbc,11103
-sarvamai/speech_to_text_translate_streaming/raw_client.py,sha256=N7BqiHl5nHIdR-fEq9B4l99aJL1UQ2CQpU6nHxM698M,10212
+sarvamai/speech_to_text_translate_streaming/client.py,sha256=d5nCU5dS9kNi2w6Eu8LiTrRn5q7-l1N-Home6oeoA0k,13255
+sarvamai/speech_to_text_translate_streaming/raw_client.py,sha256=4E_T5HepaCE80dLq_fIA4lb2mM8-feOzuC_tPuTVIbs,12364
 sarvamai/speech_to_text_translate_streaming/socket_client.py,sha256=ipEPSj5eHAyDpuEXfaP7JJL1rXJXGEo-IB888ReAFKs,8901
 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
@@ -123,12 +123,12 @@ sarvamai/text_to_speech/__init__.py,sha256=_VhToAyIt_5axN6CLJwtxg3-CO7THa_23pbUz
 sarvamai/text_to_speech/client.py,sha256=iwrQNfoMgCSOgvztTIXtLHQmSmn0RInwt5RSo9TwdtA,9617
 sarvamai/text_to_speech/raw_client.py,sha256=si_aSjMR7SocIpKZFoVYqBmaIDuRm_6vxTM0dJ73PEo,15569
 sarvamai/text_to_speech_streaming/__init__.py,sha256=TjBDNPwnsPQpuVZzh7Xi27A51pEEfVyaC8dpvafkmvQ,201
-sarvamai/text_to_speech_streaming/client.py,sha256=3WsyEcRI6h5EUeGCNZzWG-t-VJA2a7ceoGkXKBBNq7s,7415
-sarvamai/text_to_speech_streaming/raw_client.py,sha256=CHX4t6SiwEvRTflnKI61q0Z3GtjHzHb4Xz-yk2exxeU,6596
+sarvamai/text_to_speech_streaming/client.py,sha256=SaYSHYzBn3BeFRZPumXQJD1EeAYISnAmDomCPxVdyPk,7793
+sarvamai/text_to_speech_streaming/raw_client.py,sha256=E-4cVcgYA1MQRq2YkLoGt1OU3ZsYQ1wPqdQ4vjhO7ug,6974
 sarvamai/text_to_speech_streaming/socket_client.py,sha256=aun1nyc5ZYziXP0FmyvkCB0WKa_VdjA0XHYI2-Z_gCo,13973
 sarvamai/text_to_speech_streaming/types/__init__.py,sha256=fx2VJtnbh4RFUETiV2aPGslEQ0lq_us4UjDbQr_2J6I,242
 sarvamai/text_to_speech_streaming/types/text_to_speech_streaming_send_completion_event.py,sha256=cZWm6cAwNwPGF8ZADtnRev_AsgM_xj5ypg2oHHQfgZI,181
-sarvamai/types/__init__.py,sha256=hBtXC7VLxZdrBEPs7-Ym6c3fYlJJS1VxHbfb4J8CcEM,8334
+sarvamai/types/__init__.py,sha256=DvAsZu5BYk6rSQSZ02cl1Prc0tGgo0p_9L9uJjzpwWI,8416
 sarvamai/types/audio_data.py,sha256=g-sQcFXiV0uFAsUJ_TjM3OSnPoCUmNkMORRJHOl152k,1545
 sarvamai/types/audio_data_input_audio_codec.py,sha256=Xb3UEr2wTniW0z8ND4iV051YHhgFazIFOd7q7UGH3nY,572
 sarvamai/types/audio_message.py,sha256=sB4EgkWkWJzipYXobkmM9AYZTTZtCpg_ySKssUeznUE,560
@@ -143,7 +143,7 @@ sarvamai/types/chat_completion_request_system_message.py,sha256=E7YhTk1zr4u7dj_y
 sarvamai/types/chat_completion_request_user_message.py,sha256=J3WhlrfOfCCe7ugmJIfP_L9st3OFtXkIjZTSuR8O9nQ,615
 sarvamai/types/chat_completion_response_message.py,sha256=wz935eBnCkSIl0I0qMxBuH4vAUCso1aHDGReMW1VHGE,744
 sarvamai/types/choice.py,sha256=uXBCsjWP9VK3XWQWZUeI4EnU10w0G9nAfKn2tJZvxko,1244
-sarvamai/types/completion_event_flag.py,sha256=HdvjxXha9Ux5KS_Lfp7Q5eaX62eIk4bXcuJgfhGFXf4,160
+sarvamai/types/completion_event_flag.py,sha256=TzDOwE5GrC2uiVpnc9vV8y7n4uT_cr_G5FR011h_zw4,92
 sarvamai/types/completion_usage.py,sha256=xYQGlQUbKqsksuV73H-1ajjfT5M7w47eLfdWXSlrI5M,843
 sarvamai/types/config_message.py,sha256=xLD2wZcXejYrmREMd-cn38da4hKfsNPKRtyAGCW0Zcg,779
 sarvamai/types/configure_connection.py,sha256=SnSNk02gQqP8e4VB4y88jjeFQ4ClpImjGLn2ANI8cZ4,1058
@@ -179,6 +179,7 @@ sarvamai/types/language_identification_response.py,sha256=jG4ZQ6KQHCiEDqC51OniOw
 sarvamai/types/numerals_format.py,sha256=xg3lYiHcnzyFwuwRcaIteJLH_Pz6pJ9n9kTlYPEnCBU,165
 sarvamai/types/ping_signal.py,sha256=cE53FRIXlc8bSo18z6jlAnOh6DhZEMX36huWEX6X3-A,695
 sarvamai/types/reasoning_effort.py,sha256=_TBLn3rQgzJAdnKqV2g0PETbrSBZl0fPLfQ5ZE9H4Pc,164
+sarvamai/types/response_speech_state.py,sha256=SYwVP-TD2nhZ2kAXw0NzuVtkRgMjbLsuXRfKHPGxhK0,231
 sarvamai/types/response_type.py,sha256=yyk0QTIQlNa9W0Uoj_5_ey_Q3Bu8Jij5GkgR0Rt_WnU,163
 sarvamai/types/role.py,sha256=3eY01zZQKB8BSD4cFDeVjz-o2qnHJKz1vnToLqbExxs,115
 sarvamai/types/sarvam_model_ids.py,sha256=iYBMglf31KQ1iUZeAMQ-2PP9NDcyHRG7goz7O9VG8qg,124
@@ -191,14 +192,14 @@ sarvamai/types/speech_to_text_model.py,sha256=DSnGHPpIT-OyRv1bpy1xqEoLw90MTKyDgc
 sarvamai/types/speech_to_text_response.py,sha256=iWRGEJeHUFIOxeEhoCQu68njeA6lcqXbT2czV-O8Wx0,1438
 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=9xJljx_Uhhr0_DQZADPHbvEYmVCNwo9XHy7_QFPVqgY,1517
 sarvamai/types/speech_to_text_translate_job_parameters.py,sha256=fvfcyzIyT0DtcRYePDvglHH-wAhGbsi3H5G4i5nuWT8,1409
 sarvamai/types/speech_to_text_translate_language.py,sha256=yikNM-roIumVG-eqBWss93wLGudZdLPwd0i3VcXH5zo,263
 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_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=SPdYTeD1H4WQiExfs-jETyTwXblrqlsthTjUHqZEWxw,1216
 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
@@ -224,6 +225,6 @@ 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-0.1.20a3.dist-info/METADATA,sha256=m3M5A9uBZCGLv8cidbPuqOpJOyOgna77JiJAzq6eGR0,26753
-sarvamai-0.1.20a3.dist-info/WHEEL,sha256=Zb28QaM1gQi8f4VCBhsUklF61CTlNYfs9YAZn-TOGFk,88
-sarvamai-0.1.20a3.dist-info/RECORD,,
+sarvamai-0.1.22a1.dist-info/METADATA,sha256=Bg_PXPy1viWbp1gdWZiXdlRf7rBYW-Kfovmpyth1dkQ,26753
+sarvamai-0.1.22a1.dist-info/WHEEL,sha256=Zb28QaM1gQi8f4VCBhsUklF61CTlNYfs9YAZn-TOGFk,88
+sarvamai-0.1.22a1.dist-info/RECORD,,