sarvamai 0.1.8rc2__tar.gz → 0.1.8rc4__tar.gz

{sarvamai-0.1.8rc2 → sarvamai-0.1.8rc4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sarvamai
-Version: 0.1.8rc2
+Version: 0.1.8rc4
 Summary: 
 Requires-Python: >=3.8,<4.0
 Classifier: Intended Audience :: Developers

{sarvamai-0.1.8rc2 → sarvamai-0.1.8rc4}/pyproject.toml

@@ -3,7 +3,7 @@ name = "sarvamai"
 
 [tool.poetry]
 name = "sarvamai"
-version = "0.1.8rc2"
+version = "0.1.8rc4"
 description = ""
 readme = "README.md"
 authors = []

{sarvamai-0.1.8rc2 → sarvamai-0.1.8rc4}/src/sarvamai/__init__.py

@@ -154,6 +154,7 @@ from .speech_to_text_translate_streaming import (
     SpeechToTextTranslateStreamingModel,
     SpeechToTextTranslateStreamingVadSignals,
 )
+from .text_to_speech_streaming import TextToSpeechStreamingModel
 from .version import __version__
 
 __all__ = [
@@ -273,6 +274,7 @@ __all__ = [
     "TextToSpeechResponse",
     "TextToSpeechResponseParams",
     "TextToSpeechSpeaker",
+    "TextToSpeechStreamingModel",
     "TimestampsModel",
     "TimestampsModelParams",
     "TooManyRequestsError",

{sarvamai-0.1.8rc2 → sarvamai-0.1.8rc4}/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/0.1.8rc2",
+            "User-Agent": "sarvamai/0.1.8rc4",
             "X-Fern-Language": "Python",
             "X-Fern-SDK-Name": "sarvamai",
-            "X-Fern-SDK-Version": "0.1.8rc2",
+            "X-Fern-SDK-Version": "0.1.8rc4",
         }
         headers["api-subscription-key"] = self.api_subscription_key
         return headers

sarvamai-0.1.8rc4/src/sarvamai/text_to_speech_streaming/__init__.py

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

{sarvamai-0.1.8rc2 → sarvamai-0.1.8rc4}/src/sarvamai/text_to_speech_streaming/client.py

@@ -3,6 +3,7 @@
 import typing
 from contextlib import asynccontextmanager, contextmanager
 
+import httpx
 import websockets
 import websockets.sync.client as websockets_sync_client
 from ..core.api_error import ApiError
@@ -10,6 +11,7 @@ from ..core.client_wrapper import AsyncClientWrapper, SyncClientWrapper
 from ..core.request_options import RequestOptions
 from .raw_client import AsyncRawTextToSpeechStreamingClient, RawTextToSpeechStreamingClient
 from .socket_client import AsyncTextToSpeechStreamingSocketClient, TextToSpeechStreamingSocketClient
+from .types.text_to_speech_streaming_model import TextToSpeechStreamingModel
 
 
 class TextToSpeechStreamingClient:
@@ -31,6 +33,7 @@ class TextToSpeechStreamingClient:
     def connect(
         self,
         *,
+        model: typing.Optional[TextToSpeechStreamingModel] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[TextToSpeechStreamingSocketClient]:
@@ -40,6 +43,9 @@ class TextToSpeechStreamingClient:
 
         Parameters
         ----------
+        model : typing.Optional[TextToSpeechStreamingModel]
+            Text to speech model to use
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -51,6 +57,10 @@ class TextToSpeechStreamingClient:
         TextToSpeechStreamingSocketClient
         """
         ws_url = self._raw_client._client_wrapper.get_environment().production + "/text-to-speech/ws"
+        query_params = httpx.QueryParams()
+        if model is not None:
+            query_params = query_params.add("model", model)
+        ws_url = ws_url + f"?{query_params}"
         headers = self._raw_client._client_wrapper.get_headers()
         if api_subscription_key is not None:
             headers["Api-Subscription-Key"] = str(api_subscription_key)
@@ -93,6 +103,7 @@ class AsyncTextToSpeechStreamingClient:
     async def connect(
         self,
         *,
+        model: typing.Optional[TextToSpeechStreamingModel] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncTextToSpeechStreamingSocketClient]:
@@ -102,6 +113,9 @@ class AsyncTextToSpeechStreamingClient:
 
         Parameters
         ----------
+        model : typing.Optional[TextToSpeechStreamingModel]
+            Text to speech model to use
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -113,6 +127,10 @@ class AsyncTextToSpeechStreamingClient:
         AsyncTextToSpeechStreamingSocketClient
         """
         ws_url = self._raw_client._client_wrapper.get_environment().production + "/text-to-speech/ws"
+        query_params = httpx.QueryParams()
+        if model is not None:
+            query_params = query_params.add("model", model)
+        ws_url = ws_url + f"?{query_params}"
         headers = self._raw_client._client_wrapper.get_headers()
         if api_subscription_key is not None:
             headers["Api-Subscription-Key"] = str(api_subscription_key)

{sarvamai-0.1.8rc2 → sarvamai-0.1.8rc4}/src/sarvamai/text_to_speech_streaming/raw_client.py

@@ -3,12 +3,14 @@
 import typing
 from contextlib import asynccontextmanager, contextmanager
 
+import httpx
 import websockets
 import websockets.sync.client as websockets_sync_client
 from ..core.api_error import ApiError
 from ..core.client_wrapper import AsyncClientWrapper, SyncClientWrapper
 from ..core.request_options import RequestOptions
 from .socket_client import AsyncTextToSpeechStreamingSocketClient, TextToSpeechStreamingSocketClient
+from .types.text_to_speech_streaming_model import TextToSpeechStreamingModel
 
 
 class RawTextToSpeechStreamingClient:
@@ -19,6 +21,7 @@ class RawTextToSpeechStreamingClient:
     def connect(
         self,
         *,
+        model: typing.Optional[TextToSpeechStreamingModel] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[TextToSpeechStreamingSocketClient]:
@@ -28,6 +31,9 @@ class RawTextToSpeechStreamingClient:
 
         Parameters
         ----------
+        model : typing.Optional[TextToSpeechStreamingModel]
+            Text to speech model to use
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -39,6 +45,10 @@ class RawTextToSpeechStreamingClient:
         TextToSpeechStreamingSocketClient
         """
         ws_url = self._client_wrapper.get_environment().production + "/text-to-speech/ws"
+        query_params = httpx.QueryParams()
+        if model is not None:
+            query_params = query_params.add("model", model)
+        ws_url = ws_url + f"?{query_params}"
         headers = self._client_wrapper.get_headers()
         if api_subscription_key is not None:
             headers["Api-Subscription-Key"] = str(api_subscription_key)
@@ -70,6 +80,7 @@ class AsyncRawTextToSpeechStreamingClient:
     async def connect(
         self,
         *,
+        model: typing.Optional[TextToSpeechStreamingModel] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncTextToSpeechStreamingSocketClient]:
@@ -79,6 +90,9 @@ class AsyncRawTextToSpeechStreamingClient:
 
         Parameters
         ----------
+        model : typing.Optional[TextToSpeechStreamingModel]
+            Text to speech model to use
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -90,6 +104,10 @@ class AsyncRawTextToSpeechStreamingClient:
         AsyncTextToSpeechStreamingSocketClient
         """
         ws_url = self._client_wrapper.get_environment().production + "/text-to-speech/ws"
+        query_params = httpx.QueryParams()
+        if model is not None:
+            query_params = query_params.add("model", model)
+        ws_url = ws_url + f"?{query_params}"
         headers = self._client_wrapper.get_headers()
         if api_subscription_key is not None:
             headers["Api-Subscription-Key"] = str(api_subscription_key)

{sarvamai-0.1.8rc2 → sarvamai-0.1.8rc4}/src/sarvamai/text_to_speech_streaming/socket_client.py

@@ -11,8 +11,10 @@ from ..types.audio_output import AudioOutput
 from ..types.close_connection import CloseConnection
 from ..types.error_response import ErrorResponse
 from ..types.initialize_connection import InitializeConnection
+from ..types.initialize_connection_data import InitializeConnectionData
 from ..types.ping_signal import PingSignal
 from ..types.send_text import SendText
+from ..types.send_text_data import SendTextData
 
 TextToSpeechStreamingSocketClientResponse = typing.Union[AudioOutput, ErrorResponse]
 
@@ -24,6 +26,7 @@ class AsyncTextToSpeechStreamingSocketClient(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(TextToSpeechStreamingSocketClientResponse, message)  # type: ignore
 
     async def start_listening(self):
@@ -39,6 +42,11 @@ class AsyncTextToSpeechStreamingSocketClient(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(TextToSpeechStreamingSocketClientResponse, raw_message)  # type: ignore
                 self._emit(EventType.MESSAGE, parsed)
         except websockets.WebSocketException as exc:
@@ -46,32 +54,47 @@ class AsyncTextToSpeechStreamingSocketClient(EventEmitterMixin):
         finally:
             self._emit(EventType.CLOSE, None)
 
-    async def send_initialize_connection(self, message: InitializeConnection) -> None:
+    async def initialize_connection(
+        self,
+        target_language_code: str,
+        speaker: str,
+    ) -> None:
         """
-        Send a message to the websocket connection.
-        The message will be sent as a InitializeConnection.
+        Initialize the TTS connection with configuration parameters.
+
+        :param target_language_code: Target language code (e.g., 'hi-IN')
+        :param speaker: Voice speaker name (e.g., 'meera', 'arvind')
         """
+        data = InitializeConnectionData(
+            target_language_code=target_language_code,
+            speaker=speaker,
+        )
+        message = InitializeConnection(data=data)
         await self._send_model(message)
 
-    async def send_send_text(self, message: SendText) -> None:
+    async def convert(self, text: str) -> None:
         """
-        Send a message to the websocket connection.
-        The message will be sent as a SendText.
+        Send text to be converted to speech.
+
+        :param text: Text to be synthesized (1-2500 characters)
         """
+        data = SendTextData(text=text)
+        message = SendText(data=data)
         await self._send_model(message)
 
-    async def send_close_connection(self, message: CloseConnection) -> None:
+    async def flush(self) -> None:
         """
-        Send a message to the websocket connection.
-        The message will be sent as a CloseConnection.
+        Signal to flush the buffer and finalize audio output.
+        This indicates the end of text input.
         """
+        message = CloseConnection()
         await self._send_model(message)
 
-    async def send_ping_signal(self, message: PingSignal) -> None:
+    async def ping(self) -> None:
         """
-        Send a message to the websocket connection.
-        The message will be sent as a PingSignal.
+        Send ping signal to keep the WebSocket connection alive.
         """
+        message = PingSignal()
         await self._send_model(message)
 
     async def recv(self) -> TextToSpeechStreamingSocketClientResponse:
@@ -79,6 +102,7 @@ class AsyncTextToSpeechStreamingSocketClient(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(TextToSpeechStreamingSocketClientResponse, data)  # type: ignore
 
     async def _send(self, data: typing.Any) -> None:
@@ -103,6 +127,7 @@ class TextToSpeechStreamingSocketClient(EventEmitterMixin):
 
     def __iter__(self):
         for message in self._websocket:
+            message = json.loads(message) if isinstance(message, str) else message
             yield parse_obj_as(TextToSpeechStreamingSocketClientResponse, message)  # type: ignore
 
     def start_listening(self):
@@ -118,6 +143,11 @@ class TextToSpeechStreamingSocketClient(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(TextToSpeechStreamingSocketClientResponse, raw_message)  # type: ignore
                 self._emit(EventType.MESSAGE, parsed)
         except websockets.WebSocketException as exc:
@@ -125,32 +155,47 @@ class TextToSpeechStreamingSocketClient(EventEmitterMixin):
         finally:
             self._emit(EventType.CLOSE, None)
 
-    def send_initialize_connection(self, message: InitializeConnection) -> None:
+    def initialize_connection(
+        self,
+        target_language_code: str,
+        speaker: str,
+    ) -> None:
         """
-        Send a message to the websocket connection.
-        The message will be sent as a InitializeConnection.
+        Initialize the TTS connection with configuration parameters.
+
+        :param target_language_code: Target language code (e.g., 'hi-IN')
+        :param speaker: Voice speaker name (e.g., 'meera', 'arvind')
         """
+        data = InitializeConnectionData(
+            target_language_code=target_language_code,
+            speaker=speaker,
+        )
+        message = InitializeConnection(data=data)
         self._send_model(message)
 
-    def send_send_text(self, message: SendText) -> None:
+    def convert(self, text: str) -> None:
         """
-        Send a message to the websocket connection.
-        The message will be sent as a SendText.
+        Send text to be converted to speech.
+
+        :param text: Text to be synthesized (1-2500 characters)
         """
+        data = SendTextData(text=text)
+        message = SendText(data=data)
         self._send_model(message)
 
-    def send_close_connection(self, message: CloseConnection) -> None:
+    def flush(self) -> None:
         """
-        Send a message to the websocket connection.
-        The message will be sent as a CloseConnection.
+        Signal to flush the buffer and finalize audio output.
+        This indicates the end of text input.
         """
+        message = CloseConnection()
         self._send_model(message)
 
-    def send_ping_signal(self, message: PingSignal) -> None:
+    def ping(self) -> None:
         """
-        Send a message to the websocket connection.
-        The message will be sent as a PingSignal.
+        Send ping signal to keep the WebSocket connection alive.
         """
+        message = PingSignal()
         self._send_model(message)
 
     def recv(self) -> TextToSpeechStreamingSocketClientResponse:
@@ -158,6 +203,7 @@ class TextToSpeechStreamingSocketClient(EventEmitterMixin):
         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(TextToSpeechStreamingSocketClientResponse, data)  # type: ignore
 
     def _send(self, data: typing.Any) -> None:

sarvamai-0.1.8rc4/src/sarvamai/text_to_speech_streaming/types/__init__.py

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

sarvamai-0.1.8rc4/src/sarvamai/text_to_speech_streaming/types/text_to_speech_streaming_model.py

@@ -0,0 +1,5 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+TextToSpeechStreamingModel = typing.Union[typing.Literal["bulbul:v1", "bulbul:v2"], typing.Any]

sarvamai-0.1.8rc2/src/sarvamai/text_to_speech_streaming/__init__.py

@@ -1,4 +0,0 @@
-# This file was auto-generated by Fern from our API Definition.
-
-# isort: skip_file
-