sarvamai 0.1.5a12__tar.gz → 0.1.5a14__tar.gz

{sarvamai-0.1.5a12 → sarvamai-0.1.5a14}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sarvamai
-Version: 0.1.5a12
+Version: 0.1.5a14
 Summary: 
 Requires-Python: >=3.8,<4.0
 Classifier: Intended Audience :: Developers

{sarvamai-0.1.5a12 → sarvamai-0.1.5a14}/pyproject.toml

@@ -3,7 +3,7 @@ name = "sarvamai"
 
 [tool.poetry]
 name = "sarvamai"
-version = "0.1.5a12"
+version = "0.1.5a14"
 description = ""
 readme = "README.md"
 authors = []

{sarvamai-0.1.5a12 → sarvamai-0.1.5a14}/src/sarvamai/__init__.py

@@ -4,7 +4,6 @@
 
 from .types import (
     AudioData,
-    AudioDataEncoding,
     AudioMessage,
     ChatCompletionRequestAssistantMessage,
     ChatCompletionRequestMessage,
@@ -113,14 +112,22 @@ from .requests import (
     TranslationResponseParams,
     TransliterationResponseParams,
 )
-from .speech_to_text_streaming import SpeechToTextStreamingLanguageCode, SpeechToTextStreamingModel
-from .speech_to_text_translate_streaming import SpeechToTextTranslateStreamingModel
+from .speech_to_text_streaming import (
+    SpeechToTextStreamingHighVadSensitivity,
+    SpeechToTextStreamingLanguageCode,
+    SpeechToTextStreamingModel,
+    SpeechToTextStreamingVadSignals,
+)
+from .speech_to_text_translate_streaming import (
+    SpeechToTextTranslateStreamingHighVadSensitivity,
+    SpeechToTextTranslateStreamingModel,
+    SpeechToTextTranslateStreamingVadSignals,
+)
 from .version import __version__
 
 __all__ = [
     "AsyncSarvamAI",
     "AudioData",
-    "AudioDataEncoding",
     "AudioDataParams",
     "AudioMessage",
     "AudioMessageParams",
@@ -183,10 +190,12 @@ __all__ = [
     "SpeechToTextResponseData",
     "SpeechToTextResponseDataParams",
     "SpeechToTextResponseParams",
+    "SpeechToTextStreamingHighVadSensitivity",
     "SpeechToTextStreamingLanguageCode",
     "SpeechToTextStreamingModel",
     "SpeechToTextStreamingResponse",
     "SpeechToTextStreamingResponseParams",
+    "SpeechToTextStreamingVadSignals",
     "SpeechToTextTranscriptionData",
     "SpeechToTextTranscriptionDataParams",
     "SpeechToTextTranslateLanguage",
@@ -195,9 +204,11 @@ __all__ = [
     "SpeechToTextTranslateResponseData",
     "SpeechToTextTranslateResponseDataParams",
     "SpeechToTextTranslateResponseParams",
+    "SpeechToTextTranslateStreamingHighVadSensitivity",
     "SpeechToTextTranslateStreamingModel",
     "SpeechToTextTranslateStreamingResponse",
     "SpeechToTextTranslateStreamingResponseParams",
+    "SpeechToTextTranslateStreamingVadSignals",
     "SpeechToTextTranslateTranscriptionData",
     "SpeechToTextTranslateTranscriptionDataParams",
     "SpokenFormNumeralsFormat",

{sarvamai-0.1.5a12 → sarvamai-0.1.5a14}/src/sarvamai/core/client_wrapper.py

@@ -17,10 +17,10 @@ class BaseClientWrapper:
 
     def get_headers(self) -> typing.Dict[str, str]:
         headers: typing.Dict[str, str] = {
-            "User-Agent": "sarvamai/0.1.5a12",
+            "User-Agent": "sarvamai/0.1.5a14",
             "X-Fern-Language": "Python",
             "X-Fern-SDK-Name": "sarvamai",
-            "X-Fern-SDK-Version": "0.1.5a12",
+            "X-Fern-SDK-Version": "0.1.5a14",
         }
         headers["api-subscription-key"] = self.api_subscription_key
         return headers

{sarvamai-0.1.5a12 → sarvamai-0.1.5a14}/src/sarvamai/requests/audio_data.py

@@ -1,7 +1,8 @@
 # This file was auto-generated by Fern from our API Definition.
 
+import typing
+
 import typing_extensions
-from ..types.audio_data_encoding import AudioDataEncoding
 
 
 class AudioDataParams(typing_extensions.TypedDict):
@@ -15,7 +16,7 @@ class AudioDataParams(typing_extensions.TypedDict):
     Audio sample rate in Hz (16kHz preferred, 8kHz least preferred)
     """
 
-    encoding: AudioDataEncoding
+    encoding: typing.Literal["audio/wav"]
     """
     Audio encoding format
     """

sarvamai-0.1.5a14/src/sarvamai/speech_to_text_streaming/__init__.py

@@ -0,0 +1,17 @@
+# This file was auto-generated by Fern from our API Definition.
+
+# isort: skip_file
+
+from .types import (
+    SpeechToTextStreamingHighVadSensitivity,
+    SpeechToTextStreamingLanguageCode,
+    SpeechToTextStreamingModel,
+    SpeechToTextStreamingVadSignals,
+)
+
+__all__ = [
+    "SpeechToTextStreamingHighVadSensitivity",
+    "SpeechToTextStreamingLanguageCode",
+    "SpeechToTextStreamingModel",
+    "SpeechToTextStreamingVadSignals",
+]

{sarvamai-0.1.5a12 → sarvamai-0.1.5a14}/src/sarvamai/speech_to_text_streaming/client.py

@@ -11,8 +11,10 @@ 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_high_vad_sensitivity import SpeechToTextStreamingHighVadSensitivity
 from .types.speech_to_text_streaming_language_code import SpeechToTextStreamingLanguageCode
 from .types.speech_to_text_streaming_model import SpeechToTextStreamingModel
+from .types.speech_to_text_streaming_vad_signals import SpeechToTextStreamingVadSignals
 
 
 class SpeechToTextStreamingClient:
@@ -36,8 +38,8 @@ class SpeechToTextStreamingClient:
         *,
         language_code: SpeechToTextStreamingLanguageCode,
         model: typing.Optional[SpeechToTextStreamingModel] = None,
-        high_vad_sensitivity: typing.Optional[str] = None,
-        vad_signals: typing.Optional[str] = None,
+        high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
+        vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[SpeechToTextStreamingSocketClient]:
@@ -52,10 +54,10 @@ class SpeechToTextStreamingClient:
         model : typing.Optional[SpeechToTextStreamingModel]
             Speech to text model to use
 
-        high_vad_sensitivity : typing.Optional[str]
+        high_vad_sensitivity : typing.Optional[SpeechToTextStreamingHighVadSensitivity]
             Enable high VAD (Voice Activity Detection) sensitivity
 
-        vad_signals : typing.Optional[str]
+        vad_signals : typing.Optional[SpeechToTextStreamingVadSignals]
             Enable VAD signals in response
 
         api_subscription_key : typing.Optional[str]
@@ -123,8 +125,8 @@ class AsyncSpeechToTextStreamingClient:
         *,
         language_code: SpeechToTextStreamingLanguageCode,
         model: typing.Optional[SpeechToTextStreamingModel] = None,
-        high_vad_sensitivity: typing.Optional[str] = None,
-        vad_signals: typing.Optional[str] = None,
+        high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
+        vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncSpeechToTextStreamingSocketClient]:
@@ -139,10 +141,10 @@ class AsyncSpeechToTextStreamingClient:
         model : typing.Optional[SpeechToTextStreamingModel]
             Speech to text model to use
 
-        high_vad_sensitivity : typing.Optional[str]
+        high_vad_sensitivity : typing.Optional[SpeechToTextStreamingHighVadSensitivity]
             Enable high VAD (Voice Activity Detection) sensitivity
 
-        vad_signals : typing.Optional[str]
+        vad_signals : typing.Optional[SpeechToTextStreamingVadSignals]
             Enable VAD signals in response
 
         api_subscription_key : typing.Optional[str]

{sarvamai-0.1.5a12 → sarvamai-0.1.5a14}/src/sarvamai/speech_to_text_streaming/raw_client.py

@@ -10,8 +10,10 @@ 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_high_vad_sensitivity import SpeechToTextStreamingHighVadSensitivity
 from .types.speech_to_text_streaming_language_code import SpeechToTextStreamingLanguageCode
 from .types.speech_to_text_streaming_model import SpeechToTextStreamingModel
+from .types.speech_to_text_streaming_vad_signals import SpeechToTextStreamingVadSignals
 
 
 class RawSpeechToTextStreamingClient:
@@ -24,8 +26,8 @@ class RawSpeechToTextStreamingClient:
         *,
         language_code: SpeechToTextStreamingLanguageCode,
         model: typing.Optional[SpeechToTextStreamingModel] = None,
-        high_vad_sensitivity: typing.Optional[str] = None,
-        vad_signals: typing.Optional[str] = None,
+        high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
+        vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[SpeechToTextStreamingSocketClient]:
@@ -40,10 +42,10 @@ class RawSpeechToTextStreamingClient:
         model : typing.Optional[SpeechToTextStreamingModel]
             Speech to text model to use
 
-        high_vad_sensitivity : typing.Optional[str]
+        high_vad_sensitivity : typing.Optional[SpeechToTextStreamingHighVadSensitivity]
             Enable high VAD (Voice Activity Detection) sensitivity
 
-        vad_signals : typing.Optional[str]
+        vad_signals : typing.Optional[SpeechToTextStreamingVadSignals]
             Enable VAD signals in response
 
         api_subscription_key : typing.Optional[str]
@@ -100,8 +102,8 @@ class AsyncRawSpeechToTextStreamingClient:
         *,
         language_code: SpeechToTextStreamingLanguageCode,
         model: typing.Optional[SpeechToTextStreamingModel] = None,
-        high_vad_sensitivity: typing.Optional[str] = None,
-        vad_signals: typing.Optional[str] = None,
+        high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
+        vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncSpeechToTextStreamingSocketClient]:
@@ -116,10 +118,10 @@ class AsyncRawSpeechToTextStreamingClient:
         model : typing.Optional[SpeechToTextStreamingModel]
             Speech to text model to use
 
-        high_vad_sensitivity : typing.Optional[str]
+        high_vad_sensitivity : typing.Optional[SpeechToTextStreamingHighVadSensitivity]
             Enable high VAD (Voice Activity Detection) sensitivity
 
-        vad_signals : typing.Optional[str]
+        vad_signals : typing.Optional[SpeechToTextStreamingVadSignals]
             Enable VAD signals in response
 
         api_subscription_key : typing.Optional[str]

{sarvamai-0.1.5a12 → sarvamai-0.1.5a14}/src/sarvamai/speech_to_text_streaming/socket_client.py

@@ -7,6 +7,7 @@ import websockets
 import websockets.sync.connection as websockets_sync_connection
 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
 
@@ -20,6 +21,7 @@ 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
 
     async def start_listening(self):
@@ -35,6 +37,7 @@ 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
                 self._emit(EventType.MESSAGE, parsed)
         except websockets.WebSocketException as exc:
@@ -42,7 +45,19 @@ class AsyncSpeechToTextStreamingSocketClient(EventEmitterMixin):
         finally:
             self._emit(EventType.CLOSE, None)
 
-    async def send_speech_to_text_streaming_audio_message(self, message: AudioMessage) -> None:
+    async def transcribe(self, audio: str, encoding="audio/wav", sample_rate=16000):
+        """
+        Sends transcription request to the server.
+        :param audio: Base64 encoded audio data
+        :param encoding: Audio encoding format (default is "audio/wav")
+        :param sample_rate: Audio sample rate in Hz (default is 16000)
+        """
+
+        return await self._send_speech_to_text_streaming_audio_message(
+            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:
         """
         Send a message to the websocket connection.
         The message will be sent as a AudioMessage.
@@ -54,6 +69,7 @@ class AsyncSpeechToTextStreamingSocketClient(EventEmitterMixin):
         Receive a message from the websocket connection.
         """
         data = await self._websocket.recv()
+        data = json.loads(data) if isinstance(data, str) else data
         return parse_obj_as(SpeechToTextStreamingSocketClientResponse, data)  # type: ignore
 
     async def _send(self, data: typing.Any) -> None:
@@ -78,6 +94,7 @@ 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
 
     def start_listening(self):
@@ -93,6 +110,7 @@ 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
                 self._emit(EventType.MESSAGE, parsed)
         except websockets.WebSocketException as exc:
@@ -100,20 +118,32 @@ class SpeechToTextStreamingSocketClient(EventEmitterMixin):
         finally:
             self._emit(EventType.CLOSE, None)
 
-    def send_speech_to_text_streaming_audio_message(self, message: AudioMessage) -> None:
+    def transcribe(self, audio: str, encoding="audio/wav", sample_rate=16000) -> None:
         """
-        Send a message to the websocket connection.
-        The message will be sent as a AudioMessage.
+        Sends transcription request to the server.
+        :param audio: Base64 encoded audio data
+        :param encoding (Optional): Audio encoding format (default is "audio/wav")
+        :param sample_rate (Optional): Audio sample rate in Hz (default is 16000)
         """
-        self._send_model(message)
+        return self._send_speech_to_text_streaming_audio_message(
+            message=AudioMessage(audio=AudioData(data=audio, sample_rate=sample_rate, encoding=encoding))
+        )
 
     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
 
+    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.
+        """
+        self._send_model(message)
+
     def _send(self, data: typing.Any) -> None:
         """
         Send a message to the websocket connection.

sarvamai-0.1.5a14/src/sarvamai/speech_to_text_streaming/types/__init__.py

@@ -0,0 +1,15 @@
+# This file was auto-generated by Fern from our API Definition.
+
+# isort: skip_file
+
+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__ = [
+    "SpeechToTextStreamingHighVadSensitivity",
+    "SpeechToTextStreamingLanguageCode",
+    "SpeechToTextStreamingModel",
+    "SpeechToTextStreamingVadSignals",
+]

sarvamai-0.1.5a14/src/sarvamai/speech_to_text_streaming/types/speech_to_text_streaming_high_vad_sensitivity.py

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

sarvamai-0.1.5a14/src/sarvamai/speech_to_text_streaming/types/speech_to_text_streaming_vad_signals.py

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

sarvamai-0.1.5a14/src/sarvamai/speech_to_text_translate_streaming/__init__.py

@@ -0,0 +1,15 @@
+# This file was auto-generated by Fern from our API Definition.
+
+# isort: skip_file
+
+from .types import (
+    SpeechToTextTranslateStreamingHighVadSensitivity,
+    SpeechToTextTranslateStreamingModel,
+    SpeechToTextTranslateStreamingVadSignals,
+)
+
+__all__ = [
+    "SpeechToTextTranslateStreamingHighVadSensitivity",
+    "SpeechToTextTranslateStreamingModel",
+    "SpeechToTextTranslateStreamingVadSignals",
+]

{sarvamai-0.1.5a12 → sarvamai-0.1.5a14}/src/sarvamai/speech_to_text_translate_streaming/client.py

@@ -11,7 +11,11 @@ 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_high_vad_sensitivity import (
+    SpeechToTextTranslateStreamingHighVadSensitivity,
+)
 from .types.speech_to_text_translate_streaming_model import SpeechToTextTranslateStreamingModel
+from .types.speech_to_text_translate_streaming_vad_signals import SpeechToTextTranslateStreamingVadSignals
 
 
 class SpeechToTextTranslateStreamingClient:
@@ -34,8 +38,8 @@ class SpeechToTextTranslateStreamingClient:
         self,
         *,
         model: typing.Optional[SpeechToTextTranslateStreamingModel] = None,
-        high_vad_sensitivity: typing.Optional[str] = None,
-        vad_signals: typing.Optional[str] = None,
+        high_vad_sensitivity: typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity] = None,
+        vad_signals: typing.Optional[SpeechToTextTranslateStreamingVadSignals] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[SpeechToTextTranslateStreamingSocketClient]:
@@ -47,10 +51,10 @@ class SpeechToTextTranslateStreamingClient:
         model : typing.Optional[SpeechToTextTranslateStreamingModel]
             Speech to text model to use (defaults to "saaras:v2.5" if not specified)
 
-        high_vad_sensitivity : typing.Optional[str]
+        high_vad_sensitivity : typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity]
             Enable high VAD (Voice Activity Detection) sensitivity
 
-        vad_signals : typing.Optional[str]
+        vad_signals : typing.Optional[SpeechToTextTranslateStreamingVadSignals]
             Enable VAD signals in response
 
         api_subscription_key : typing.Optional[str]
@@ -115,8 +119,8 @@ class AsyncSpeechToTextTranslateStreamingClient:
         self,
         *,
         model: typing.Optional[SpeechToTextTranslateStreamingModel] = None,
-        high_vad_sensitivity: typing.Optional[str] = None,
-        vad_signals: typing.Optional[str] = None,
+        high_vad_sensitivity: typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity] = None,
+        vad_signals: typing.Optional[SpeechToTextTranslateStreamingVadSignals] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncSpeechToTextTranslateStreamingSocketClient]:
@@ -128,10 +132,10 @@ class AsyncSpeechToTextTranslateStreamingClient:
         model : typing.Optional[SpeechToTextTranslateStreamingModel]
             Speech to text model to use (defaults to "saaras:v2.5" if not specified)
 
-        high_vad_sensitivity : typing.Optional[str]
+        high_vad_sensitivity : typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity]
             Enable high VAD (Voice Activity Detection) sensitivity
 
-        vad_signals : typing.Optional[str]
+        vad_signals : typing.Optional[SpeechToTextTranslateStreamingVadSignals]
             Enable VAD signals in response
 
         api_subscription_key : typing.Optional[str]

{sarvamai-0.1.5a12 → sarvamai-0.1.5a14}/src/sarvamai/speech_to_text_translate_streaming/raw_client.py

@@ -10,7 +10,11 @@ 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_high_vad_sensitivity import (
+    SpeechToTextTranslateStreamingHighVadSensitivity,
+)
 from .types.speech_to_text_translate_streaming_model import SpeechToTextTranslateStreamingModel
+from .types.speech_to_text_translate_streaming_vad_signals import SpeechToTextTranslateStreamingVadSignals
 
 
 class RawSpeechToTextTranslateStreamingClient:
@@ -22,8 +26,8 @@ class RawSpeechToTextTranslateStreamingClient:
         self,
         *,
         model: typing.Optional[SpeechToTextTranslateStreamingModel] = None,
-        high_vad_sensitivity: typing.Optional[str] = None,
-        vad_signals: typing.Optional[str] = None,
+        high_vad_sensitivity: typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity] = None,
+        vad_signals: typing.Optional[SpeechToTextTranslateStreamingVadSignals] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[SpeechToTextTranslateStreamingSocketClient]:
@@ -35,10 +39,10 @@ class RawSpeechToTextTranslateStreamingClient:
         model : typing.Optional[SpeechToTextTranslateStreamingModel]
             Speech to text model to use (defaults to "saaras:v2.5" if not specified)
 
-        high_vad_sensitivity : typing.Optional[str]
+        high_vad_sensitivity : typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity]
             Enable high VAD (Voice Activity Detection) sensitivity
 
-        vad_signals : typing.Optional[str]
+        vad_signals : typing.Optional[SpeechToTextTranslateStreamingVadSignals]
             Enable VAD signals in response
 
         api_subscription_key : typing.Optional[str]
@@ -92,8 +96,8 @@ class AsyncRawSpeechToTextTranslateStreamingClient:
         self,
         *,
         model: typing.Optional[SpeechToTextTranslateStreamingModel] = None,
-        high_vad_sensitivity: typing.Optional[str] = None,
-        vad_signals: typing.Optional[str] = None,
+        high_vad_sensitivity: typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity] = None,
+        vad_signals: typing.Optional[SpeechToTextTranslateStreamingVadSignals] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncSpeechToTextTranslateStreamingSocketClient]:
@@ -105,10 +109,10 @@ class AsyncRawSpeechToTextTranslateStreamingClient:
         model : typing.Optional[SpeechToTextTranslateStreamingModel]
             Speech to text model to use (defaults to "saaras:v2.5" if not specified)
 
-        high_vad_sensitivity : typing.Optional[str]
+        high_vad_sensitivity : typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity]
             Enable high VAD (Voice Activity Detection) sensitivity
 
-        vad_signals : typing.Optional[str]
+        vad_signals : typing.Optional[SpeechToTextTranslateStreamingVadSignals]
             Enable VAD signals in response
 
         api_subscription_key : typing.Optional[str]