sarvamai 0.1.18a0__py3-none-any.whl → 0.1.19a1__py3-none-any.whl

sarvamai/__init__.py

@@ -74,6 +74,7 @@ from .types import (
     SpokenFormNumeralsFormat,
     StopConfiguration,
     StorageContainerType,
+    SttFlushSignal,
     TaskDetailV1,
     TaskFileDetails,
     TaskState,
@@ -167,6 +168,7 @@ from .requests import (
     SpeechToTextTranslateStreamingResponseParams,
     SpeechToTextTranslateTranscriptionDataParams,
     StopConfigurationParams,
+    SttFlushSignalParams,
     TaskDetailV1Params,
     TaskFileDetailsParams,
     TextToSpeechResponseParams,
@@ -176,12 +178,14 @@ from .requests import (
     TransliterationResponseParams,
 )
 from .speech_to_text_streaming import (
+    SpeechToTextStreamingFlushSignal,
     SpeechToTextStreamingHighVadSensitivity,
     SpeechToTextStreamingLanguageCode,
     SpeechToTextStreamingModel,
     SpeechToTextStreamingVadSignals,
 )
 from .speech_to_text_translate_streaming import (
+    SpeechToTextTranslateStreamingFlushSignal,
     SpeechToTextTranslateStreamingHighVadSensitivity,
     SpeechToTextTranslateStreamingModel,
     SpeechToTextTranslateStreamingVadSignals,
@@ -297,6 +301,7 @@ __all__ = [
     "SpeechToTextResponseData",
     "SpeechToTextResponseDataParams",
     "SpeechToTextResponseParams",
+    "SpeechToTextStreamingFlushSignal",
     "SpeechToTextStreamingHighVadSensitivity",
     "SpeechToTextStreamingLanguageCode",
     "SpeechToTextStreamingModel",
@@ -313,6 +318,7 @@ __all__ = [
     "SpeechToTextTranslateResponseData",
     "SpeechToTextTranslateResponseDataParams",
     "SpeechToTextTranslateResponseParams",
+    "SpeechToTextTranslateStreamingFlushSignal",
     "SpeechToTextTranslateStreamingHighVadSensitivity",
     "SpeechToTextTranslateStreamingModel",
     "SpeechToTextTranslateStreamingResponse",
@@ -324,6 +330,8 @@ __all__ = [
     "StopConfiguration",
     "StopConfigurationParams",
     "StorageContainerType",
+    "SttFlushSignal",
+    "SttFlushSignalParams",
     "TaskDetailV1",
     "TaskDetailV1Params",
     "TaskFileDetails",

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/0.1.18a0",
+            "User-Agent": "sarvamai/0.1.19a1",
             "X-Fern-Language": "Python",
             "X-Fern-SDK-Name": "sarvamai",
-            "X-Fern-SDK-Version": "0.1.18a0",
+            "X-Fern-SDK-Version": "0.1.19a1",
             **(self.get_custom_headers() or {}),
         }
         headers["api-subscription-key"] = self.api_subscription_key

sarvamai/requests/__init__.py

@@ -54,6 +54,7 @@ from .speech_to_text_translate_response_data import SpeechToTextTranslateRespons
 from .speech_to_text_translate_streaming_response import SpeechToTextTranslateStreamingResponseParams
 from .speech_to_text_translate_transcription_data import SpeechToTextTranslateTranscriptionDataParams
 from .stop_configuration import StopConfigurationParams
+from .stt_flush_signal import SttFlushSignalParams
 from .task_detail_v_1 import TaskDetailV1Params
 from .task_file_details import TaskFileDetailsParams
 from .text_to_speech_response import TextToSpeechResponseParams
@@ -113,6 +114,7 @@ __all__ = [
     "SpeechToTextTranslateStreamingResponseParams",
     "SpeechToTextTranslateTranscriptionDataParams",
     "StopConfigurationParams",
+    "SttFlushSignalParams",
     "TaskDetailV1Params",
     "TaskFileDetailsParams",
     "TextToSpeechResponseParams",

sarvamai/requests/stt_flush_signal.py

@@ -0,0 +1,21 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+import typing_extensions
+
+
+class SttFlushSignalParams(typing_extensions.TypedDict):
+    """
+    Signal to flush the audio buffer and force finalize partial transcriptions/translations
+    """
+
+    type: typing.Literal["flush"]
+    """
+    Type identifier for flush signal
+    """
+
+    reason: typing_extensions.NotRequired[str]
+    """
+    Optional reason for flushing (e.g., "end_of_segment", "manual_flush")
+    """

sarvamai/speech_to_text_streaming/__init__.py

@@ -3,6 +3,7 @@
 # isort: skip_file
 
 from .types import (
+    SpeechToTextStreamingFlushSignal,
     SpeechToTextStreamingHighVadSensitivity,
     SpeechToTextStreamingLanguageCode,
     SpeechToTextStreamingModel,
@@ -10,6 +11,7 @@ from .types import (
 )
 
 __all__ = [
+    "SpeechToTextStreamingFlushSignal",
     "SpeechToTextStreamingHighVadSensitivity",
     "SpeechToTextStreamingLanguageCode",
     "SpeechToTextStreamingModel",

sarvamai/speech_to_text_streaming/client.py

@@ -11,6 +11,7 @@ from ..core.client_wrapper import AsyncClientWrapper, SyncClientWrapper
 from ..core.request_options import RequestOptions
 from .raw_client import AsyncRawSpeechToTextStreamingClient, RawSpeechToTextStreamingClient
 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_language_code import SpeechToTextStreamingLanguageCode
 from .types.speech_to_text_streaming_model import SpeechToTextStreamingModel
@@ -45,6 +46,7 @@ class SpeechToTextStreamingClient:
         model: typing.Optional[SpeechToTextStreamingModel] = None,
         high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
+        flush_signal: typing.Optional[SpeechToTextStreamingFlushSignal] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[SpeechToTextStreamingSocketClient]:
@@ -65,6 +67,9 @@ class SpeechToTextStreamingClient:
         vad_signals : typing.Optional[SpeechToTextStreamingVadSignals]
             Enable VAD signals in response
 
+        flush_signal : typing.Optional[SpeechToTextStreamingFlushSignal]
+            Signal to flush the audio buffer and finalize transcription
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -85,6 +90,8 @@ class SpeechToTextStreamingClient:
             query_params = query_params.add("high_vad_sensitivity", high_vad_sensitivity)
         if vad_signals is not None:
             query_params = query_params.add("vad_signals", vad_signals)
+        if flush_signal is not None:
+            query_params = query_params.add("flush_signal", flush_signal)
         ws_url = ws_url + f"?{query_params}"
         headers = self._raw_client._client_wrapper.get_headers()
         if api_subscription_key is not None:
@@ -132,6 +139,7 @@ class AsyncSpeechToTextStreamingClient:
         model: typing.Optional[SpeechToTextStreamingModel] = None,
         high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
+        flush_signal: typing.Optional[SpeechToTextStreamingFlushSignal] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncSpeechToTextStreamingSocketClient]:
@@ -152,6 +160,9 @@ class AsyncSpeechToTextStreamingClient:
         vad_signals : typing.Optional[SpeechToTextStreamingVadSignals]
             Enable VAD signals in response
 
+        flush_signal : typing.Optional[SpeechToTextStreamingFlushSignal]
+            Signal to flush the audio buffer and finalize transcription
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -172,6 +183,8 @@ class AsyncSpeechToTextStreamingClient:
             query_params = query_params.add("high_vad_sensitivity", high_vad_sensitivity)
         if vad_signals is not None:
             query_params = query_params.add("vad_signals", vad_signals)
+        if flush_signal is not None:
+            query_params = query_params.add("flush_signal", flush_signal)
         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

@@ -10,6 +10,7 @@ from ..core.api_error import ApiError
 from ..core.client_wrapper import AsyncClientWrapper, SyncClientWrapper
 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_language_code import SpeechToTextStreamingLanguageCode
 from .types.speech_to_text_streaming_model import SpeechToTextStreamingModel
@@ -33,6 +34,7 @@ class RawSpeechToTextStreamingClient:
         model: typing.Optional[SpeechToTextStreamingModel] = None,
         high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
+        flush_signal: typing.Optional[SpeechToTextStreamingFlushSignal] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[SpeechToTextStreamingSocketClient]:
@@ -53,6 +55,9 @@ class RawSpeechToTextStreamingClient:
         vad_signals : typing.Optional[SpeechToTextStreamingVadSignals]
             Enable VAD signals in response
 
+        flush_signal : typing.Optional[SpeechToTextStreamingFlushSignal]
+            Signal to flush the audio buffer and finalize transcription
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -73,6 +78,8 @@ class RawSpeechToTextStreamingClient:
             query_params = query_params.add("high_vad_sensitivity", high_vad_sensitivity)
         if vad_signals is not None:
             query_params = query_params.add("vad_signals", vad_signals)
+        if flush_signal is not None:
+            query_params = query_params.add("flush_signal", flush_signal)
         ws_url = ws_url + f"?{query_params}"
         headers = self._client_wrapper.get_headers()
         if api_subscription_key is not None:
@@ -109,6 +116,7 @@ class AsyncRawSpeechToTextStreamingClient:
         model: typing.Optional[SpeechToTextStreamingModel] = None,
         high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
+        flush_signal: typing.Optional[SpeechToTextStreamingFlushSignal] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncSpeechToTextStreamingSocketClient]:
@@ -129,6 +137,9 @@ class AsyncRawSpeechToTextStreamingClient:
         vad_signals : typing.Optional[SpeechToTextStreamingVadSignals]
             Enable VAD signals in response
 
+        flush_signal : typing.Optional[SpeechToTextStreamingFlushSignal]
+            Signal to flush the audio buffer and finalize transcription
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -149,6 +160,8 @@ class AsyncRawSpeechToTextStreamingClient:
             query_params = query_params.add("high_vad_sensitivity", high_vad_sensitivity)
         if vad_signals is not None:
             query_params = query_params.add("vad_signals", vad_signals)
+        if flush_signal is not None:
+            query_params = query_params.add("flush_signal", flush_signal)
         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/socket_client.py

@@ -9,7 +9,10 @@ from ..core.events import EventEmitterMixin, EventType
 from ..core.pydantic_utilities import parse_obj_as
 from ..types.audio_data import AudioData
 from ..types.audio_message import AudioMessage
-from ..types.speech_to_text_streaming_response import SpeechToTextStreamingResponse
+from ..types.speech_to_text_streaming_response import (
+    SpeechToTextStreamingResponse,
+)
+from ..types.stt_flush_signal import SttFlushSignal
 
 SpeechToTextStreamingSocketClientResponse = typing.Union[SpeechToTextStreamingResponse]
 
@@ -22,7 +25,9 @@ class AsyncSpeechToTextStreamingSocketClient(EventEmitterMixin):
     async def __aiter__(self):
         async for message in self._websocket:
             message = json.loads(message) if isinstance(message, str) else message
-            yield parse_obj_as(SpeechToTextStreamingSocketClientResponse, message)  # type: ignore
+            yield parse_obj_as(
+                SpeechToTextStreamingSocketClientResponse, message
+            )  # type: ignore
 
     async def start_listening(self):
         """
@@ -37,8 +42,14 @@ class AsyncSpeechToTextStreamingSocketClient(EventEmitterMixin):
         self._emit(EventType.OPEN, None)
         try:
             async for raw_message in self._websocket:
-                raw_message = json.loads(raw_message) if isinstance(raw_message, str) else raw_message
-                parsed = parse_obj_as(SpeechToTextStreamingSocketClientResponse, raw_message)  # type: ignore
+                raw_message = (
+                    json.loads(raw_message)
+                    if isinstance(raw_message, str)
+                    else raw_message
+                )
+                parsed = parse_obj_as(
+                    SpeechToTextStreamingSocketClientResponse, raw_message
+                )  # type: ignore
                 self._emit(EventType.MESSAGE, parsed)
         except websockets.WebSocketException as exc:
             self._emit(EventType.ERROR, exc)
@@ -54,10 +65,26 @@ class AsyncSpeechToTextStreamingSocketClient(EventEmitterMixin):
         """
 
         return await self._send_speech_to_text_streaming_audio_message(
-            message=AudioMessage(audio=AudioData(data=audio, sample_rate=sample_rate, encoding=encoding))
+            message=AudioMessage(
+                audio=AudioData(data=audio, sample_rate=sample_rate, encoding=encoding)
+            )
         )
 
-    async def _send_speech_to_text_streaming_audio_message(self, message: AudioMessage) -> None:
+    async def flush(self, reason: typing.Optional[str] = None) -> None:
+        """
+        Signal to flush the audio buffer and force finalize partial transcriptions.
+        Use this to force processing of any remaining audio that hasn't been
+        transcribed yet.
+
+        :param reason: Optional reason for flushing (e.g., "end_of_segment",
+                      "manual_flush")
+        """
+        message = SttFlushSignal(reason=reason)
+        await self._send_model(message)
+
+    async def _send_speech_to_text_streaming_audio_message(
+        self, message: AudioMessage
+    ) -> None:
         """
         Send a message to the websocket connection.
         The message will be sent as a AudioMessage.
@@ -70,7 +97,9 @@ class AsyncSpeechToTextStreamingSocketClient(EventEmitterMixin):
         """
         data = await self._websocket.recv()
         data = json.loads(data) if isinstance(data, str) else data
-        return parse_obj_as(SpeechToTextStreamingSocketClientResponse, data)  # type: ignore
+        return parse_obj_as(
+            SpeechToTextStreamingSocketClientResponse, data
+        )  # type: ignore
 
     async def _send(self, data: typing.Any) -> None:
         """
@@ -95,7 +124,9 @@ class SpeechToTextStreamingSocketClient(EventEmitterMixin):
     def __iter__(self):
         for message in self._websocket:
             message = json.loads(message) if isinstance(message, str) else message
-            yield parse_obj_as(SpeechToTextStreamingSocketClientResponse, message)  # type: ignore
+            yield parse_obj_as(
+                SpeechToTextStreamingSocketClientResponse, message
+            )  # type: ignore
 
     def start_listening(self):
         """
@@ -110,8 +141,14 @@ class SpeechToTextStreamingSocketClient(EventEmitterMixin):
         self._emit(EventType.OPEN, None)
         try:
             for raw_message in self._websocket:
-                raw_message = json.loads(raw_message) if isinstance(raw_message, str) else raw_message
-                parsed = parse_obj_as(SpeechToTextStreamingSocketClientResponse, raw_message)  # type: ignore
+                raw_message = (
+                    json.loads(raw_message)
+                    if isinstance(raw_message, str)
+                    else raw_message
+                )
+                parsed = parse_obj_as(
+                    SpeechToTextStreamingSocketClientResponse, raw_message
+                )  # type: ignore
                 self._emit(EventType.MESSAGE, parsed)
         except websockets.WebSocketException as exc:
             self._emit(EventType.ERROR, exc)
@@ -126,18 +163,36 @@ class SpeechToTextStreamingSocketClient(EventEmitterMixin):
         :param sample_rate (Optional): Audio sample rate in Hz (default is 16000)
         """
         return self._send_speech_to_text_streaming_audio_message(
-            message=AudioMessage(audio=AudioData(data=audio, sample_rate=sample_rate, encoding=encoding))
+            message=AudioMessage(
+                audio=AudioData(data=audio, sample_rate=sample_rate, encoding=encoding)
+            )
         )
 
+    def flush(self, reason: typing.Optional[str] = None) -> None:
+        """
+        Signal to flush the audio buffer and force finalize partial transcriptions.
+        Use this to force processing of any remaining audio that hasn't been
+        transcribed yet.
+
+        :param reason: Optional reason for flushing (e.g., "end_of_segment",
+                      "manual_flush")
+        """
+        message = SttFlushSignal(reason=reason)
+        self._send_model(message)
+
     def recv(self) -> SpeechToTextStreamingSocketClientResponse:
         """
         Receive a message from the websocket connection.
         """
         data = self._websocket.recv()
         data = json.loads(data) if isinstance(data, str) else data
-        return parse_obj_as(SpeechToTextStreamingSocketClientResponse, data)  # type: ignore
+        return parse_obj_as(
+            SpeechToTextStreamingSocketClientResponse, data
+        )  # type: ignore
 
-    def _send_speech_to_text_streaming_audio_message(self, message: AudioMessage) -> None:
+    def _send_speech_to_text_streaming_audio_message(
+        self, message: AudioMessage
+    ) -> None:
         """
         Send a message to the websocket connection.
         The message will be sent as a AudioMessage.

sarvamai/speech_to_text_streaming/types/__init__.py

@@ -2,12 +2,14 @@
 
 # isort: skip_file
 
+from .speech_to_text_streaming_flush_signal import SpeechToTextStreamingFlushSignal
 from .speech_to_text_streaming_high_vad_sensitivity import SpeechToTextStreamingHighVadSensitivity
 from .speech_to_text_streaming_language_code import SpeechToTextStreamingLanguageCode
 from .speech_to_text_streaming_model import SpeechToTextStreamingModel
 from .speech_to_text_streaming_vad_signals import SpeechToTextStreamingVadSignals
 
 __all__ = [
+    "SpeechToTextStreamingFlushSignal",
     "SpeechToTextStreamingHighVadSensitivity",
     "SpeechToTextStreamingLanguageCode",
     "SpeechToTextStreamingModel",

sarvamai/speech_to_text_streaming/types/speech_to_text_streaming_flush_signal.py

@@ -0,0 +1,5 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+SpeechToTextStreamingFlushSignal = typing.Union[typing.Literal["true", "false"], typing.Any]

sarvamai/speech_to_text_translate_streaming/__init__.py

@@ -3,12 +3,14 @@
 # isort: skip_file
 
 from .types import (
+    SpeechToTextTranslateStreamingFlushSignal,
     SpeechToTextTranslateStreamingHighVadSensitivity,
     SpeechToTextTranslateStreamingModel,
     SpeechToTextTranslateStreamingVadSignals,
 )
 
 __all__ = [
+    "SpeechToTextTranslateStreamingFlushSignal",
     "SpeechToTextTranslateStreamingHighVadSensitivity",
     "SpeechToTextTranslateStreamingModel",
     "SpeechToTextTranslateStreamingVadSignals",

sarvamai/speech_to_text_translate_streaming/client.py

@@ -11,6 +11,7 @@ from ..core.client_wrapper import AsyncClientWrapper, SyncClientWrapper
 from ..core.request_options import RequestOptions
 from .raw_client import AsyncRawSpeechToTextTranslateStreamingClient, RawSpeechToTextTranslateStreamingClient
 from .socket_client import AsyncSpeechToTextTranslateStreamingSocketClient, SpeechToTextTranslateStreamingSocketClient
+from .types.speech_to_text_translate_streaming_flush_signal import SpeechToTextTranslateStreamingFlushSignal
 from .types.speech_to_text_translate_streaming_high_vad_sensitivity import (
     SpeechToTextTranslateStreamingHighVadSensitivity,
 )
@@ -45,6 +46,7 @@ class SpeechToTextTranslateStreamingClient:
         model: typing.Optional[SpeechToTextTranslateStreamingModel] = None,
         high_vad_sensitivity: typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextTranslateStreamingVadSignals] = None,
+        flush_signal: typing.Optional[SpeechToTextTranslateStreamingFlushSignal] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[SpeechToTextTranslateStreamingSocketClient]:
@@ -62,6 +64,9 @@ class SpeechToTextTranslateStreamingClient:
         vad_signals : typing.Optional[SpeechToTextTranslateStreamingVadSignals]
             Enable VAD signals in response
 
+        flush_signal : typing.Optional[SpeechToTextTranslateStreamingFlushSignal]
+            Signal to flush the audio buffer and finalize transcription and translation
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -80,6 +85,8 @@ class SpeechToTextTranslateStreamingClient:
             query_params = query_params.add("high_vad_sensitivity", high_vad_sensitivity)
         if vad_signals is not None:
             query_params = query_params.add("vad_signals", vad_signals)
+        if flush_signal is not None:
+            query_params = query_params.add("flush_signal", flush_signal)
         ws_url = ws_url + f"?{query_params}"
         headers = self._raw_client._client_wrapper.get_headers()
         if api_subscription_key is not None:
@@ -126,6 +133,7 @@ class AsyncSpeechToTextTranslateStreamingClient:
         model: typing.Optional[SpeechToTextTranslateStreamingModel] = None,
         high_vad_sensitivity: typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextTranslateStreamingVadSignals] = None,
+        flush_signal: typing.Optional[SpeechToTextTranslateStreamingFlushSignal] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncSpeechToTextTranslateStreamingSocketClient]:
@@ -143,6 +151,9 @@ class AsyncSpeechToTextTranslateStreamingClient:
         vad_signals : typing.Optional[SpeechToTextTranslateStreamingVadSignals]
             Enable VAD signals in response
 
+        flush_signal : typing.Optional[SpeechToTextTranslateStreamingFlushSignal]
+            Signal to flush the audio buffer and finalize transcription and translation
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -161,6 +172,8 @@ class AsyncSpeechToTextTranslateStreamingClient:
             query_params = query_params.add("high_vad_sensitivity", high_vad_sensitivity)
         if vad_signals is not None:
             query_params = query_params.add("vad_signals", vad_signals)
+        if flush_signal is not None:
+            query_params = query_params.add("flush_signal", flush_signal)
         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

@@ -10,6 +10,7 @@ from ..core.api_error import ApiError
 from ..core.client_wrapper import AsyncClientWrapper, SyncClientWrapper
 from ..core.request_options import RequestOptions
 from .socket_client import AsyncSpeechToTextTranslateStreamingSocketClient, SpeechToTextTranslateStreamingSocketClient
+from .types.speech_to_text_translate_streaming_flush_signal import SpeechToTextTranslateStreamingFlushSignal
 from .types.speech_to_text_translate_streaming_high_vad_sensitivity import (
     SpeechToTextTranslateStreamingHighVadSensitivity,
 )
@@ -33,6 +34,7 @@ class RawSpeechToTextTranslateStreamingClient:
         model: typing.Optional[SpeechToTextTranslateStreamingModel] = None,
         high_vad_sensitivity: typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextTranslateStreamingVadSignals] = None,
+        flush_signal: typing.Optional[SpeechToTextTranslateStreamingFlushSignal] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[SpeechToTextTranslateStreamingSocketClient]:
@@ -50,6 +52,9 @@ class RawSpeechToTextTranslateStreamingClient:
         vad_signals : typing.Optional[SpeechToTextTranslateStreamingVadSignals]
             Enable VAD signals in response
 
+        flush_signal : typing.Optional[SpeechToTextTranslateStreamingFlushSignal]
+            Signal to flush the audio buffer and finalize transcription and translation
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -68,6 +73,8 @@ class RawSpeechToTextTranslateStreamingClient:
             query_params = query_params.add("high_vad_sensitivity", high_vad_sensitivity)
         if vad_signals is not None:
             query_params = query_params.add("vad_signals", vad_signals)
+        if flush_signal is not None:
+            query_params = query_params.add("flush_signal", flush_signal)
         ws_url = ws_url + f"?{query_params}"
         headers = self._client_wrapper.get_headers()
         if api_subscription_key is not None:
@@ -103,6 +110,7 @@ class AsyncRawSpeechToTextTranslateStreamingClient:
         model: typing.Optional[SpeechToTextTranslateStreamingModel] = None,
         high_vad_sensitivity: typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextTranslateStreamingVadSignals] = None,
+        flush_signal: typing.Optional[SpeechToTextTranslateStreamingFlushSignal] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncSpeechToTextTranslateStreamingSocketClient]:
@@ -120,6 +128,9 @@ class AsyncRawSpeechToTextTranslateStreamingClient:
         vad_signals : typing.Optional[SpeechToTextTranslateStreamingVadSignals]
             Enable VAD signals in response
 
+        flush_signal : typing.Optional[SpeechToTextTranslateStreamingFlushSignal]
+            Signal to flush the audio buffer and finalize transcription and translation
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -138,6 +149,8 @@ class AsyncRawSpeechToTextTranslateStreamingClient:
             query_params = query_params.add("high_vad_sensitivity", high_vad_sensitivity)
         if vad_signals is not None:
             query_params = query_params.add("vad_signals", vad_signals)
+        if flush_signal is not None:
+            query_params = query_params.add("flush_signal", flush_signal)
         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

@@ -2,11 +2,13 @@
 
 # isort: skip_file
 
+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_model import SpeechToTextTranslateStreamingModel
 from .speech_to_text_translate_streaming_vad_signals import SpeechToTextTranslateStreamingVadSignals
 
 __all__ = [
+    "SpeechToTextTranslateStreamingFlushSignal",
     "SpeechToTextTranslateStreamingHighVadSensitivity",
     "SpeechToTextTranslateStreamingModel",
     "SpeechToTextTranslateStreamingVadSignals",

sarvamai/speech_to_text_translate_streaming/types/speech_to_text_translate_streaming_flush_signal.py

@@ -0,0 +1,5 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+SpeechToTextTranslateStreamingFlushSignal = typing.Union[typing.Literal["true", "false"], typing.Any]

sarvamai/types/__init__.py

@@ -75,6 +75,7 @@ from .speech_to_text_translate_transcription_data import SpeechToTextTranslateTr
 from .spoken_form_numerals_format import SpokenFormNumeralsFormat
 from .stop_configuration import StopConfiguration
 from .storage_container_type import StorageContainerType
+from .stt_flush_signal import SttFlushSignal
 from .task_detail_v_1 import TaskDetailV1
 from .task_file_details import TaskFileDetails
 from .task_state import TaskState
@@ -168,6 +169,7 @@ __all__ = [
     "SpokenFormNumeralsFormat",
     "StopConfiguration",
     "StorageContainerType",
+    "SttFlushSignal",
     "TaskDetailV1",
     "TaskFileDetails",
     "TaskState",

sarvamai/types/stt_flush_signal.py

@@ -0,0 +1,31 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+import pydantic
+from ..core.pydantic_utilities import IS_PYDANTIC_V2, UniversalBaseModel
+
+
+class SttFlushSignal(UniversalBaseModel):
+    """
+    Signal to flush the audio buffer and force finalize partial transcriptions/translations
+    """
+
+    type: typing.Literal["flush"] = pydantic.Field(default="flush")
+    """
+    Type identifier for flush signal
+    """
+
+    reason: typing.Optional[str] = pydantic.Field(default=None)
+    """
+    Optional reason for flushing (e.g., "end_of_segment", "manual_flush")
+    """
+
+    if IS_PYDANTIC_V2:
+        model_config: typing.ClassVar[pydantic.ConfigDict] = pydantic.ConfigDict(extra="allow", frozen=True)  # type: ignore # Pydantic v2
+    else:
+
+        class Config:
+            frozen = True
+            smart_union = True
+            extra = pydantic.Extra.allow

{sarvamai-0.1.18a0.dist-info → sarvamai-0.1.19a1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sarvamai
-Version: 0.1.18a0
+Version: 0.1.19a1
 Summary: 
 Requires-Python: >=3.8,<4.0
 Classifier: Intended Audience :: Developers

{sarvamai-0.1.18a0.dist-info → sarvamai-0.1.19a1.dist-info}/RECORD

@@ -1,11 +1,11 @@
-sarvamai/__init__.py,sha256=5j3UOszpw8O8ihXP2HwhHPgwCGnzPu5-8w9FDV51k1s,10694
+sarvamai/__init__.py,sha256=R2X_oWI35aSTjAskLMeKysc5rt17-ldLPc8vhkfGku4,10964
 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=aSFsptwfVLJJ4_L-Hjc7FiBWuK9kfyLo-rQlIcN5mn4,2570
+sarvamai/core/client_wrapper.py,sha256=zeaP6DuMbs7G4AqP-w_QTivx8CVyHsw-Dycgh5yhdiI,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
@@ -28,7 +28,7 @@ sarvamai/errors/too_many_requests_error.py,sha256=Dl-_pfpboXJh-OtSbRaPQOB-UXvpVO
 sarvamai/errors/unprocessable_entity_error.py,sha256=JqxtzIhvjkpQDqbT9Q-go1n-gyv9PsYqq0ng_ZYyBMo,347
 sarvamai/play.py,sha256=4fh86zy8g8IPU2O8yPBY7QxXQOivv_nWQvPQsOa1arw,2183
 sarvamai/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-sarvamai/requests/__init__.py,sha256=ahxnrLbGzmGrHbwbacfelPA6m2OAXy4dlLFQiF5_BNQ,5623
+sarvamai/requests/__init__.py,sha256=nAXyZKbjwhLTxvfb1kaxOxZpyFvWwQu30Yr5HqgOZXw,5702
 sarvamai/requests/audio_data.py,sha256=QI3SK5aiAg2yJ-m3l9CxOkONnH3CCKMFCl9kAdMs19o,410
 sarvamai/requests/audio_message.py,sha256=ZBeogjGE6YFXXM-0g8zq9SoizDk21reR0YXSB-0fMjg,214
 sarvamai/requests/audio_output.py,sha256=BnoX345rwoWgaMaj24u_19-SjmPV0xt7vlFEEDKRw20,280
@@ -76,6 +76,7 @@ sarvamai/requests/speech_to_text_translate_response_data.py,sha256=OmjunP9R2xert
 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/stop_configuration.py,sha256=Xmp8zyUpnN65pH5A7NqefckB8wk53_BBzOUrgRm2gXs,146
+sarvamai/requests/stt_flush_signal.py,sha256=dVb3oD3ZyXKyFd08O9O5PXJEx20oBx7vkUW5crFtxVg,498
 sarvamai/requests/task_detail_v_1.py,sha256=2rehl7dSDSgzaw13b9bODamhiN2uB-IK4cOksq8Vmqc,582
 sarvamai/requests/task_file_details.py,sha256=yf6NVXymizG9wTsCGl-xolsxDA_nDOzM3qeabiqMs60,186
 sarvamai/requests/text_to_speech_response.py,sha256=-FNMERV6zrwIH27htwBy3QiPY8WhbTmlmQ06RegYQaw,391
@@ -90,11 +91,12 @@ sarvamai/speech_to_text_job/__init__.py,sha256=_VhToAyIt_5axN6CLJwtxg3-CO7THa_23
 sarvamai/speech_to_text_job/client.py,sha256=WSGBJxYcNxl77Zd1X6VVWjg4zshqecXf6WCyhfLXVlI,18007
 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=q7QygMmZCHJ-4FMhhL_6XNV_dsqlIFRCO1iSxoyxaaY,437
-sarvamai/speech_to_text_streaming/client.py,sha256=AzStfZDXhu2YAJEpnVbsy0WObub5ctlGBzqfeYOUlpA,8442
-sarvamai/speech_to_text_streaming/raw_client.py,sha256=UrSRTR6NtB6SU0Ta8U2ER-PxBP1sDWL857l2LIaTxiU,7623
-sarvamai/speech_to_text_streaming/socket_client.py,sha256=QI0vEjDvNTG9-EH99NolIuARYwD-P8Fms8rqKOD9HJQ,6419
-sarvamai/speech_to_text_streaming/types/__init__.py,sha256=hzEDbcyy6K0Q9-Zx5k5lxAHxIgeHnAoPpvTLrL13YT0,599
+sarvamai/speech_to_text_streaming/__init__.py,sha256=aVcI8kSlUpwLrPtA-4oUkuJ_D9Onha_qU49lA2vtEqU,515
+sarvamai/speech_to_text_streaming/client.py,sha256=15k-UH2ROia2EymXia0BxVKbzBz7k1Rh9k7L9hgO5Fg,9206
+sarvamai/speech_to_text_streaming/raw_client.py,sha256=69hAAD9mcpplo3gR7B5JXbo6Yys8bPT3DfSk1FjyI8Q,8387
+sarvamai/speech_to_text_streaming/socket_client.py,sha256=N0wjuQ37WtTggQysWfB8pOwLIwmIkGRglZO9N2y5yHI,7870
+sarvamai/speech_to_text_streaming/types/__init__.py,sha256=A99VJwG5fkik5SFTMDMKPZCjvKSaU2aQ02EbA5rrDlM,723
+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_model.py,sha256=b6F4ymgz4got6KVDqrweYvkET8itze63wUwWyjqDlO4,180
@@ -103,11 +105,12 @@ 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=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=_hmlce1Zs1grylysZhBUdtKfkaUROwVydtwz6l-1qqg,411
-sarvamai/speech_to_text_translate_streaming/client.py,sha256=xPPg5_JgpH8tYDUte6FGtpzXO2LGBUSRADN-ICqqA6U,8286
-sarvamai/speech_to_text_translate_streaming/raw_client.py,sha256=WUj5dGWQT2HUYLFoi3SRJzXk1JMuLxhPQzbL45G5kO8,7395
+sarvamai/speech_to_text_translate_streaming/__init__.py,sha256=sDeWLagKUTg34tL7vpNNwr2gX4AznEn6NAy9jY-Znf4,507
+sarvamai/speech_to_text_translate_streaming/client.py,sha256=41FAJekotqq08tDdxWqhT966B5Ofx9MuZgHAJ9xmwRQ,9137
+sarvamai/speech_to_text_translate_streaming/raw_client.py,sha256=P-p7pNJNLrgra4HoUx1vG34RcLADYfQEpoVMYe-aMa0,8246
 sarvamai/speech_to_text_translate_streaming/socket_client.py,sha256=Xy86G9Mv1kGLf8xwFo3OBf98vqGQ50iigUAoOaApLdI,7842
-sarvamai/speech_to_text_translate_streaming/types/__init__.py,sha256=zyKoGAbKW4d0-Zi56F0RfPUqtk_xUjWjF_RjTxbXXW4,556
+sarvamai/speech_to_text_translate_streaming/types/__init__.py,sha256=sK4Zv64ZcV33opFFaVd1kgnQnzjfXrSDhUwiEM85sZc,708
+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_model.py,sha256=6B8VxkpJG_pNprCSctseDtJb_ULVdKrPaeENkQ6Jvjg,187
 sarvamai/speech_to_text_translate_streaming/types/speech_to_text_translate_streaming_vad_signals.py,sha256=EV3xd9qyKMnMvA9rO-qFDDIac4b84roBu7n-maaPxG8,181
@@ -121,7 +124,7 @@ sarvamai/text_to_speech_streaming/__init__.py,sha256=_VhToAyIt_5axN6CLJwtxg3-CO7
 sarvamai/text_to_speech_streaming/client.py,sha256=tcYTUmWGseGQkLOsJdO4S3eZrFstjaLCPi5OcKflfus,6306
 sarvamai/text_to_speech_streaming/raw_client.py,sha256=VdL5HYpUotIm5HaOqTwRkyZqoV0BuzwCtAfcB0g1y10,5487
 sarvamai/text_to_speech_streaming/socket_client.py,sha256=NEcijnvjuNcWfzqpBi-xWsXVkL0NPq6EGAkEjnaq9hw,13909
-sarvamai/types/__init__.py,sha256=ilRhyVqS8gsWX_73IclMel-MzGS_zL5jk-7VrsdlmDU,7862
+sarvamai/types/__init__.py,sha256=Zydymb6llnD7V1vFWM42RrXvyxvtE84e5ayeXL5iV30,7929
 sarvamai/types/audio_data.py,sha256=rgOukLkLNJ_HBBVE2g5dfEL2CWjRoGiMvCtpq0qTB1Y,829
 sarvamai/types/audio_message.py,sha256=sB4EgkWkWJzipYXobkmM9AYZTTZtCpg_ySKssUeznUE,560
 sarvamai/types/audio_output.py,sha256=Eq-YUZa1mSDwt7bax2c4Vv2gBlyM_JBJWzHhTAhFSko,621
@@ -190,6 +193,7 @@ sarvamai/types/speech_to_text_translate_transcription_data.py,sha256=-cZZm21um6e
 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
+sarvamai/types/stt_flush_signal.py,sha256=L-xRXxVRUzZ5uJGyf4-Y7yl5VZH8DF-lIIh31Z_cM10,892
 sarvamai/types/task_detail_v_1.py,sha256=Z-RotwB9BzegZqur--4HkCafyTcDvr0CosZ52St0Eeo,857
 sarvamai/types/task_file_details.py,sha256=oJV7zaUVrbqqw-If-2_V1aLk28qW0ZbeIDtIpn5-xUM,544
 sarvamai/types/task_state.py,sha256=fSrmD00Goi0J6s9hzqcFqz3Fkh37diBYpxnz4FkwHdU,182
@@ -211,6 +215,6 @@ 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-0.1.18a0.dist-info/METADATA,sha256=GCcWgRfMGBERxdaob_qr3shLcTHkF88vYGMoUC2_AMc,26753
-sarvamai-0.1.18a0.dist-info/WHEEL,sha256=Zb28QaM1gQi8f4VCBhsUklF61CTlNYfs9YAZn-TOGFk,88
-sarvamai-0.1.18a0.dist-info/RECORD,,
+sarvamai-0.1.19a1.dist-info/METADATA,sha256=x59PbuWdgTKrL5EAZfr0SlbC0jqQKP5BfP0fGNjKQPE,26753
+sarvamai-0.1.19a1.dist-info/WHEEL,sha256=Zb28QaM1gQi8f4VCBhsUklF61CTlNYfs9YAZn-TOGFk,88
+sarvamai-0.1.19a1.dist-info/RECORD,,