sarvamai 0.1.5a10__tar.gz → 0.1.5a13__tar.gz

{sarvamai-0.1.5a10 → sarvamai-0.1.5a13}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sarvamai
-Version: 0.1.5a10
+Version: 0.1.5a13
 Summary: 
 Requires-Python: >=3.8,<4.0
 Classifier: Intended Audience :: Developers

{sarvamai-0.1.5a10 → sarvamai-0.1.5a13}/pyproject.toml

@@ -3,7 +3,7 @@ name = "sarvamai"
 
 [tool.poetry]
 name = "sarvamai"
-version = "0.1.5a10"
+version = "0.1.5a13"
 description = ""
 readme = "README.md"
 authors = []

{sarvamai-0.1.5a10 → sarvamai-0.1.5a13}/src/sarvamai/__init__.py

@@ -16,6 +16,7 @@ from .types import (
     ChatCompletionResponseMessage,
     Choice,
     CompletionUsage,
+    ConfigMessage,
     CreateChatCompletionResponse,
     DiarizedEntry,
     DiarizedTranscript,
@@ -29,18 +30,22 @@ from .types import (
     LanguageIdentificationResponse,
     NumeralsFormat,
     ReasoningEffort,
+    ResponseType,
     Role,
     SarvamModelIds,
     SpeechSampleRate,
     SpeechToTextLanguage,
     SpeechToTextModel,
     SpeechToTextResponse,
+    SpeechToTextResponseData,
     SpeechToTextStreamingResponse,
-    SpeechToTextStreamingResponseData,
-    SpeechToTextStreamingResponseType,
+    SpeechToTextTranscriptionData,
     SpeechToTextTranslateLanguage,
     SpeechToTextTranslateModel,
     SpeechToTextTranslateResponse,
+    SpeechToTextTranslateResponseData,
+    SpeechToTextTranslateStreamingResponse,
+    SpeechToTextTranslateTranscriptionData,
     SpokenFormNumeralsFormat,
     StopConfiguration,
     TextToSpeechLanguage,
@@ -48,7 +53,6 @@ from .types import (
     TextToSpeechResponse,
     TextToSpeechSpeaker,
     TimestampsModel,
-    TranscriptionData,
     TranscriptionMetrics,
     TranslateMode,
     TranslateModel,
@@ -69,7 +73,7 @@ from .errors import (
     TooManyRequestsError,
     UnprocessableEntityError,
 )
-from . import chat, speech_to_text, speech_to_text_streaming, text, text_to_speech
+from . import chat, speech_to_text, speech_to_text_streaming, speech_to_text_translate_streaming, text, text_to_speech
 from .client import AsyncSarvamAI, SarvamAI
 from .environment import SarvamAIEnvironment
 from .requests import (
@@ -85,6 +89,7 @@ from .requests import (
     ChatCompletionResponseMessageParams,
     ChoiceParams,
     CompletionUsageParams,
+    ConfigMessageParams,
     CreateChatCompletionResponseParams,
     DiarizedEntryParams,
     DiarizedTranscriptParams,
@@ -93,19 +98,23 @@ from .requests import (
     ErrorMessageParams,
     EventsDataParams,
     LanguageIdentificationResponseParams,
+    SpeechToTextResponseDataParams,
     SpeechToTextResponseParams,
-    SpeechToTextStreamingResponseDataParams,
     SpeechToTextStreamingResponseParams,
+    SpeechToTextTranscriptionDataParams,
+    SpeechToTextTranslateResponseDataParams,
     SpeechToTextTranslateResponseParams,
+    SpeechToTextTranslateStreamingResponseParams,
+    SpeechToTextTranslateTranscriptionDataParams,
     StopConfigurationParams,
     TextToSpeechResponseParams,
     TimestampsModelParams,
-    TranscriptionDataParams,
     TranscriptionMetricsParams,
     TranslationResponseParams,
     TransliterationResponseParams,
 )
 from .speech_to_text_streaming import SpeechToTextStreamingLanguageCode, SpeechToTextStreamingModel
+from .speech_to_text_translate_streaming import SpeechToTextTranslateStreamingModel
 from .version import __version__
 
 __all__ = [
@@ -136,6 +145,8 @@ __all__ = [
     "ChoiceParams",
     "CompletionUsage",
     "CompletionUsageParams",
+    "ConfigMessage",
+    "ConfigMessageParams",
     "CreateChatCompletionResponse",
     "CreateChatCompletionResponseParams",
     "DiarizedEntry",
@@ -159,6 +170,7 @@ __all__ = [
     "LanguageIdentificationResponseParams",
     "NumeralsFormat",
     "ReasoningEffort",
+    "ResponseType",
     "Role",
     "SarvamAI",
     "SarvamAIEnvironment",
@@ -168,18 +180,26 @@ __all__ = [
     "SpeechToTextLanguage",
     "SpeechToTextModel",
     "SpeechToTextResponse",
+    "SpeechToTextResponseData",
+    "SpeechToTextResponseDataParams",
     "SpeechToTextResponseParams",
     "SpeechToTextStreamingLanguageCode",
     "SpeechToTextStreamingModel",
     "SpeechToTextStreamingResponse",
-    "SpeechToTextStreamingResponseData",
-    "SpeechToTextStreamingResponseDataParams",
     "SpeechToTextStreamingResponseParams",
-    "SpeechToTextStreamingResponseType",
+    "SpeechToTextTranscriptionData",
+    "SpeechToTextTranscriptionDataParams",
     "SpeechToTextTranslateLanguage",
     "SpeechToTextTranslateModel",
     "SpeechToTextTranslateResponse",
+    "SpeechToTextTranslateResponseData",
+    "SpeechToTextTranslateResponseDataParams",
     "SpeechToTextTranslateResponseParams",
+    "SpeechToTextTranslateStreamingModel",
+    "SpeechToTextTranslateStreamingResponse",
+    "SpeechToTextTranslateStreamingResponseParams",
+    "SpeechToTextTranslateTranscriptionData",
+    "SpeechToTextTranslateTranscriptionDataParams",
     "SpokenFormNumeralsFormat",
     "StopConfiguration",
     "StopConfigurationParams",
@@ -191,8 +211,6 @@ __all__ = [
     "TimestampsModel",
     "TimestampsModelParams",
     "TooManyRequestsError",
-    "TranscriptionData",
-    "TranscriptionDataParams",
     "TranscriptionMetrics",
     "TranscriptionMetricsParams",
     "TranslateMode",
@@ -212,6 +230,7 @@ __all__ = [
     "chat",
     "speech_to_text",
     "speech_to_text_streaming",
+    "speech_to_text_translate_streaming",
     "text",
     "text_to_speech",
 ]

{sarvamai-0.1.5a10 → sarvamai-0.1.5a13}/src/sarvamai/client.py

@@ -10,6 +10,10 @@ from .core.client_wrapper import AsyncClientWrapper, SyncClientWrapper
 from .environment import SarvamAIEnvironment
 from .speech_to_text.client import AsyncSpeechToTextClient, SpeechToTextClient
 from .speech_to_text_streaming.client import AsyncSpeechToTextStreamingClient, SpeechToTextStreamingClient
+from .speech_to_text_translate_streaming.client import (
+    AsyncSpeechToTextTranslateStreamingClient,
+    SpeechToTextTranslateStreamingClient,
+)
 from .text.client import AsyncTextClient, TextClient
 from .text_to_speech.client import AsyncTextToSpeechClient, TextToSpeechClient
 
@@ -79,6 +83,9 @@ class SarvamAI:
         self.text_to_speech = TextToSpeechClient(client_wrapper=self._client_wrapper)
         self.chat = ChatClient(client_wrapper=self._client_wrapper)
         self.speech_to_text_streaming = SpeechToTextStreamingClient(client_wrapper=self._client_wrapper)
+        self.speech_to_text_translate_streaming = SpeechToTextTranslateStreamingClient(
+            client_wrapper=self._client_wrapper
+        )
 
 
 class AsyncSarvamAI:
@@ -146,3 +153,6 @@ class AsyncSarvamAI:
         self.text_to_speech = AsyncTextToSpeechClient(client_wrapper=self._client_wrapper)
         self.chat = AsyncChatClient(client_wrapper=self._client_wrapper)
         self.speech_to_text_streaming = AsyncSpeechToTextStreamingClient(client_wrapper=self._client_wrapper)
+        self.speech_to_text_translate_streaming = AsyncSpeechToTextTranslateStreamingClient(
+            client_wrapper=self._client_wrapper
+        )

{sarvamai-0.1.5a10 → sarvamai-0.1.5a13}/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/0.1.5a10",
+            "User-Agent": "sarvamai/0.1.5a13",
             "X-Fern-Language": "Python",
             "X-Fern-SDK-Name": "sarvamai",
-            "X-Fern-SDK-Version": "0.1.5a10",
+            "X-Fern-SDK-Version": "0.1.5a13",
         }
         headers["api-subscription-key"] = self.api_subscription_key
         return headers

{sarvamai-0.1.5a10 → sarvamai-0.1.5a13}/src/sarvamai/requests/__init__.py

@@ -16,6 +16,7 @@ from .chat_completion_request_user_message import ChatCompletionRequestUserMessa
 from .chat_completion_response_message import ChatCompletionResponseMessageParams
 from .choice import ChoiceParams
 from .completion_usage import CompletionUsageParams
+from .config_message import ConfigMessageParams
 from .create_chat_completion_response import CreateChatCompletionResponseParams
 from .diarized_entry import DiarizedEntryParams
 from .diarized_transcript import DiarizedTranscriptParams
@@ -25,13 +26,16 @@ from .error_message import ErrorMessageParams
 from .events_data import EventsDataParams
 from .language_identification_response import LanguageIdentificationResponseParams
 from .speech_to_text_response import SpeechToTextResponseParams
+from .speech_to_text_response_data import SpeechToTextResponseDataParams
 from .speech_to_text_streaming_response import SpeechToTextStreamingResponseParams
-from .speech_to_text_streaming_response_data import SpeechToTextStreamingResponseDataParams
+from .speech_to_text_transcription_data import SpeechToTextTranscriptionDataParams
 from .speech_to_text_translate_response import SpeechToTextTranslateResponseParams
+from .speech_to_text_translate_response_data import SpeechToTextTranslateResponseDataParams
+from .speech_to_text_translate_streaming_response import SpeechToTextTranslateStreamingResponseParams
+from .speech_to_text_translate_transcription_data import SpeechToTextTranslateTranscriptionDataParams
 from .stop_configuration import StopConfigurationParams
 from .text_to_speech_response import TextToSpeechResponseParams
 from .timestamps_model import TimestampsModelParams
-from .transcription_data import TranscriptionDataParams
 from .transcription_metrics import TranscriptionMetricsParams
 from .translation_response import TranslationResponseParams
 from .transliteration_response import TransliterationResponseParams
@@ -49,6 +53,7 @@ __all__ = [
     "ChatCompletionResponseMessageParams",
     "ChoiceParams",
     "CompletionUsageParams",
+    "ConfigMessageParams",
     "CreateChatCompletionResponseParams",
     "DiarizedEntryParams",
     "DiarizedTranscriptParams",
@@ -57,14 +62,17 @@ __all__ = [
     "ErrorMessageParams",
     "EventsDataParams",
     "LanguageIdentificationResponseParams",
+    "SpeechToTextResponseDataParams",
     "SpeechToTextResponseParams",
-    "SpeechToTextStreamingResponseDataParams",
     "SpeechToTextStreamingResponseParams",
+    "SpeechToTextTranscriptionDataParams",
+    "SpeechToTextTranslateResponseDataParams",
     "SpeechToTextTranslateResponseParams",
+    "SpeechToTextTranslateStreamingResponseParams",
+    "SpeechToTextTranslateTranscriptionDataParams",
     "StopConfigurationParams",
     "TextToSpeechResponseParams",
     "TimestampsModelParams",
-    "TranscriptionDataParams",
     "TranscriptionMetricsParams",
     "TranslationResponseParams",
     "TransliterationResponseParams",

sarvamai-0.1.5a13/src/sarvamai/requests/config_message.py

@@ -0,0 +1,17 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+import typing_extensions
+
+
+class ConfigMessageParams(typing_extensions.TypedDict):
+    type: typing.Literal["config"]
+    """
+    Message type identifier for configuration
+    """
+
+    prompt: typing_extensions.NotRequired[str]
+    """
+    Prompt for ASR model to improve transcription accuracy
+    """

sarvamai-0.1.5a13/src/sarvamai/requests/speech_to_text_response_data.py

@@ -0,0 +1,9 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+from .error_data import ErrorDataParams
+from .events_data import EventsDataParams
+from .speech_to_text_transcription_data import SpeechToTextTranscriptionDataParams
+
+SpeechToTextResponseDataParams = typing.Union[SpeechToTextTranscriptionDataParams, ErrorDataParams, EventsDataParams]

sarvamai-0.1.5a13/src/sarvamai/requests/speech_to_text_streaming_response.py

@@ -0,0 +1,10 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing_extensions
+from ..types.response_type import ResponseType
+from .speech_to_text_response_data import SpeechToTextResponseDataParams
+
+
+class SpeechToTextStreamingResponseParams(typing_extensions.TypedDict):
+    type: ResponseType
+    data: SpeechToTextResponseDataParams

sarvamai-0.1.5a10/src/sarvamai/requests/transcription_data.py → sarvamai-0.1.5a13/src/sarvamai/requests/speech_to_text_transcription_data.py

@@ -6,7 +6,7 @@ import typing_extensions
 from .transcription_metrics import TranscriptionMetricsParams
 
 
-class TranscriptionDataParams(typing_extensions.TypedDict):
+class SpeechToTextTranscriptionDataParams(typing_extensions.TypedDict):
     request_id: str
     """
     Unique identifier for the request
@@ -14,7 +14,7 @@ class TranscriptionDataParams(typing_extensions.TypedDict):
 
     transcript: str
     """
-    Transcript of the provided speech
+    Transcript of the provided speech in original language
     """
 
     timestamps: typing_extensions.NotRequired[typing.Dict[str, typing.Optional[typing.Any]]]

sarvamai-0.1.5a13/src/sarvamai/requests/speech_to_text_translate_response_data.py

@@ -0,0 +1,11 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+from .error_data import ErrorDataParams
+from .events_data import EventsDataParams
+from .speech_to_text_translate_transcription_data import SpeechToTextTranslateTranscriptionDataParams
+
+SpeechToTextTranslateResponseDataParams = typing.Union[
+    SpeechToTextTranslateTranscriptionDataParams, ErrorDataParams, EventsDataParams
+]

sarvamai-0.1.5a13/src/sarvamai/requests/speech_to_text_translate_streaming_response.py

@@ -0,0 +1,10 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing_extensions
+from ..types.response_type import ResponseType
+from .speech_to_text_translate_response_data import SpeechToTextTranslateResponseDataParams
+
+
+class SpeechToTextTranslateStreamingResponseParams(typing_extensions.TypedDict):
+    type: ResponseType
+    data: SpeechToTextTranslateResponseDataParams

sarvamai-0.1.5a13/src/sarvamai/requests/speech_to_text_translate_transcription_data.py

@@ -0,0 +1,23 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing_extensions
+from .transcription_metrics import TranscriptionMetricsParams
+
+
+class SpeechToTextTranslateTranscriptionDataParams(typing_extensions.TypedDict):
+    request_id: str
+    """
+    Unique identifier for the request
+    """
+
+    transcript: str
+    """
+    English translation of the provided speech
+    """
+
+    language_code: typing_extensions.NotRequired[str]
+    """
+    BCP-47 code of detected source language (null when language detection is in progress)
+    """
+
+    metrics: TranscriptionMetricsParams

{sarvamai-0.1.5a10 → sarvamai-0.1.5a13}/src/sarvamai/speech_to_text_streaming/client.py

@@ -35,7 +35,7 @@ class SpeechToTextStreamingClient:
         self,
         *,
         language_code: SpeechToTextStreamingLanguageCode,
-        model: SpeechToTextStreamingModel,
+        model: typing.Optional[SpeechToTextStreamingModel] = None,
         high_vad_sensitivity: typing.Optional[str] = None,
         vad_signals: typing.Optional[str] = None,
         api_subscription_key: typing.Optional[str] = None,
@@ -49,7 +49,7 @@ class SpeechToTextStreamingClient:
         language_code : SpeechToTextStreamingLanguageCode
             Language code for speech recognition
 
-        model : SpeechToTextStreamingModel
+        model : typing.Optional[SpeechToTextStreamingModel]
             Speech to text model to use
 
         high_vad_sensitivity : typing.Optional[str]
@@ -122,7 +122,7 @@ class AsyncSpeechToTextStreamingClient:
         self,
         *,
         language_code: SpeechToTextStreamingLanguageCode,
-        model: SpeechToTextStreamingModel,
+        model: typing.Optional[SpeechToTextStreamingModel] = None,
         high_vad_sensitivity: typing.Optional[str] = None,
         vad_signals: typing.Optional[str] = None,
         api_subscription_key: typing.Optional[str] = None,
@@ -136,7 +136,7 @@ class AsyncSpeechToTextStreamingClient:
         language_code : SpeechToTextStreamingLanguageCode
             Language code for speech recognition
 
-        model : SpeechToTextStreamingModel
+        model : typing.Optional[SpeechToTextStreamingModel]
             Speech to text model to use
 
         high_vad_sensitivity : typing.Optional[str]

{sarvamai-0.1.5a10 → sarvamai-0.1.5a13}/src/sarvamai/speech_to_text_streaming/raw_client.py

@@ -23,7 +23,7 @@ class RawSpeechToTextStreamingClient:
         self,
         *,
         language_code: SpeechToTextStreamingLanguageCode,
-        model: SpeechToTextStreamingModel,
+        model: typing.Optional[SpeechToTextStreamingModel] = None,
         high_vad_sensitivity: typing.Optional[str] = None,
         vad_signals: typing.Optional[str] = None,
         api_subscription_key: typing.Optional[str] = None,
@@ -37,7 +37,7 @@ class RawSpeechToTextStreamingClient:
         language_code : SpeechToTextStreamingLanguageCode
             Language code for speech recognition
 
-        model : SpeechToTextStreamingModel
+        model : typing.Optional[SpeechToTextStreamingModel]
             Speech to text model to use
 
         high_vad_sensitivity : typing.Optional[str]
@@ -99,7 +99,7 @@ class AsyncRawSpeechToTextStreamingClient:
         self,
         *,
         language_code: SpeechToTextStreamingLanguageCode,
-        model: SpeechToTextStreamingModel,
+        model: typing.Optional[SpeechToTextStreamingModel] = None,
         high_vad_sensitivity: typing.Optional[str] = None,
         vad_signals: typing.Optional[str] = None,
         api_subscription_key: typing.Optional[str] = None,
@@ -113,7 +113,7 @@ class AsyncRawSpeechToTextStreamingClient:
         language_code : SpeechToTextStreamingLanguageCode
             Language code for speech recognition
 
-        model : SpeechToTextStreamingModel
+        model : typing.Optional[SpeechToTextStreamingModel]
             Speech to text model to use
 
         high_vad_sensitivity : typing.Optional[str]

{sarvamai-0.1.5a10 → sarvamai-0.1.5a13}/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
 
@@ -44,7 +45,19 @@ class AsyncSpeechToTextStreamingSocketClient(EventEmitterMixin):
         finally:
             self._emit(EventType.CLOSE, None)
 
-    async def send_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.
@@ -81,6 +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):
@@ -104,12 +118,16 @@ class SpeechToTextStreamingSocketClient(EventEmitterMixin):
         finally:
             self._emit(EventType.CLOSE, None)
 
-    def send_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:
         """
@@ -119,6 +137,13 @@ class SpeechToTextStreamingSocketClient(EventEmitterMixin):
         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-0.1.5a13/src/sarvamai/speech_to_text_translate_streaming/__init__.py

@@ -0,0 +1,7 @@
+# This file was auto-generated by Fern from our API Definition.
+
+# isort: skip_file
+
+from .types import SpeechToTextTranslateStreamingModel
+
+__all__ = ["SpeechToTextTranslateStreamingModel"]