sarvamai 0.1.8rc4__tar.gz → 0.1.8rc6__tar.gz

{sarvamai-0.1.8rc4 → sarvamai-0.1.8rc6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sarvamai
-Version: 0.1.8rc4
+Version: 0.1.8rc6
 Summary: 
 Requires-Python: >=3.8,<4.0
 Classifier: Intended Audience :: Developers

{sarvamai-0.1.8rc4 → sarvamai-0.1.8rc6}/pyproject.toml

@@ -3,7 +3,7 @@ name = "sarvamai"
 
 [tool.poetry]
 name = "sarvamai"
-version = "0.1.8rc4"
+version = "0.1.8rc6"
 description = ""
 readme = "README.md"
 authors = []

{sarvamai-0.1.8rc4 → sarvamai-0.1.8rc6}/src/sarvamai/__init__.py

@@ -16,7 +16,6 @@ from .types import (
     ChatCompletionRequestUserMessage,
     ChatCompletionResponseMessage,
     Choice,
-    CloseConnection,
     CompletionUsage,
     ConfigMessage,
     CreateChatCompletionResponse,
@@ -30,6 +29,7 @@ from .types import (
     ErrorResponseData,
     EventsData,
     FinishReason,
+    FlushSignal,
     Format,
     InitializeConnection,
     InitializeConnectionData,
@@ -110,7 +110,6 @@ from .requests import (
     ChatCompletionRequestUserMessageParams,
     ChatCompletionResponseMessageParams,
     ChoiceParams,
-    CloseConnectionParams,
     CompletionUsageParams,
     ConfigMessageParams,
     CreateChatCompletionResponseParams,
@@ -122,6 +121,7 @@ from .requests import (
     ErrorResponseDataParams,
     ErrorResponseParams,
     EventsDataParams,
+    FlushSignalParams,
     InitializeConnectionDataParams,
     InitializeConnectionParams,
     LanguageIdentificationResponseParams,
@@ -154,7 +154,6 @@ from .speech_to_text_translate_streaming import (
     SpeechToTextTranslateStreamingModel,
     SpeechToTextTranslateStreamingVadSignals,
 )
-from .text_to_speech_streaming import TextToSpeechStreamingModel
 from .version import __version__
 
 __all__ = [
@@ -186,8 +185,6 @@ __all__ = [
     "ChatCompletionResponseMessageParams",
     "Choice",
     "ChoiceParams",
-    "CloseConnection",
-    "CloseConnectionParams",
     "CompletionUsage",
     "CompletionUsageParams",
     "ConfigMessage",
@@ -212,6 +209,8 @@ __all__ = [
     "EventsData",
     "EventsDataParams",
     "FinishReason",
+    "FlushSignal",
+    "FlushSignalParams",
     "ForbiddenError",
     "Format",
     "InitializeConnection",
@@ -274,7 +273,6 @@ __all__ = [
     "TextToSpeechResponse",
     "TextToSpeechResponseParams",
     "TextToSpeechSpeaker",
-    "TextToSpeechStreamingModel",
     "TimestampsModel",
     "TimestampsModelParams",
     "TooManyRequestsError",

{sarvamai-0.1.8rc4 → sarvamai-0.1.8rc6}/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/0.1.8rc4",
+            "User-Agent": "sarvamai/0.1.8rc6",
             "X-Fern-Language": "Python",
             "X-Fern-SDK-Name": "sarvamai",
-            "X-Fern-SDK-Version": "0.1.8rc4",
+            "X-Fern-SDK-Version": "0.1.8rc6",
         }
         headers["api-subscription-key"] = self.api_subscription_key
         return headers

{sarvamai-0.1.8rc4 → sarvamai-0.1.8rc6}/src/sarvamai/requests/__init__.py

@@ -17,7 +17,6 @@ from .chat_completion_request_system_message import ChatCompletionRequestSystemM
 from .chat_completion_request_user_message import ChatCompletionRequestUserMessageParams
 from .chat_completion_response_message import ChatCompletionResponseMessageParams
 from .choice import ChoiceParams
-from .close_connection import CloseConnectionParams
 from .completion_usage import CompletionUsageParams
 from .config_message import ConfigMessageParams
 from .create_chat_completion_response import CreateChatCompletionResponseParams
@@ -29,6 +28,7 @@ from .error_message import ErrorMessageParams
 from .error_response import ErrorResponseParams
 from .error_response_data import ErrorResponseDataParams
 from .events_data import EventsDataParams
+from .flush_signal import FlushSignalParams
 from .initialize_connection import InitializeConnectionParams
 from .initialize_connection_data import InitializeConnectionDataParams
 from .language_identification_response import LanguageIdentificationResponseParams
@@ -64,7 +64,6 @@ __all__ = [
     "ChatCompletionRequestUserMessageParams",
     "ChatCompletionResponseMessageParams",
     "ChoiceParams",
-    "CloseConnectionParams",
     "CompletionUsageParams",
     "ConfigMessageParams",
     "CreateChatCompletionResponseParams",
@@ -76,6 +75,7 @@ __all__ = [
     "ErrorResponseDataParams",
     "ErrorResponseParams",
     "EventsDataParams",
+    "FlushSignalParams",
     "InitializeConnectionDataParams",
     "InitializeConnectionParams",
     "LanguageIdentificationResponseParams",

sarvamai-0.1.8rc6/src/sarvamai/requests/flush_signal.py

@@ -0,0 +1,14 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+import typing_extensions
+
+
+class FlushSignalParams(typing_extensions.TypedDict):
+    """
+    Forces the text buffer to process immediately, regardless of the min_buffer_size threshold.
+    Use this when you need to process remaining text that hasn't reached the minimum buffer size.
+    """
+
+    type: typing.Literal["flush"]

sarvamai-0.1.8rc6/src/sarvamai/requests/initialize_connection.py

@@ -0,0 +1,18 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+import typing_extensions
+from .initialize_connection_data import InitializeConnectionDataParams
+
+
+class InitializeConnectionParams(typing_extensions.TypedDict):
+    """
+    Configuration message required as the first message after establishing the WebSocket connection.
+    This initializes TTS parameters and can be updated at any time during the WebSocket lifecycle
+    by sending a new config message. When a config update is sent, any text currently in the buffer
+    will be automatically flushed and processed before applying the new configuration.
+    """
+
+    type: typing.Literal["config"]
+    data: InitializeConnectionDataParams

sarvamai-0.1.8rc6/src/sarvamai/requests/initialize_connection_data.py

@@ -0,0 +1,83 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+import typing_extensions
+from ..types.initialize_connection_data_output_audio_bitrate import InitializeConnectionDataOutputAudioBitrate
+from ..types.initialize_connection_data_speaker import InitializeConnectionDataSpeaker
+from ..types.initialize_connection_data_target_language_code import InitializeConnectionDataTargetLanguageCode
+
+
+class InitializeConnectionDataParams(typing_extensions.TypedDict):
+    target_language_code: InitializeConnectionDataTargetLanguageCode
+    """
+    The language of the text is BCP-47 format
+    """
+
+    speaker: InitializeConnectionDataSpeaker
+    """
+    The speaker voice to be used for the output audio.
+    
+    **Default:** Anushka
+    
+    **Model Compatibility (Speakers compatible with respective model):**
+    - **bulbul:v2:**
+      - Female: Anushka, Manisha, Vidya, Arya
+      - Male: Abhilash, Karun, Hitesh
+    
+    **Note:** Speaker selection must match the chosen model version.
+    """
+
+    pitch: typing_extensions.NotRequired[float]
+    """
+    Controls the pitch of the audio. Lower values result in a deeper voice, 
+    while higher values make it sharper. The suitable range is between -0.75 
+    and 0.75. Default is 0.0.
+    """
+
+    pace: typing_extensions.NotRequired[float]
+    """
+    Controls the speed of the audio. Lower values result in slower speech, 
+    while higher values make it faster. The suitable range is between 0.5 
+    and 2.0. Default is 1.0.
+    """
+
+    loudness: typing_extensions.NotRequired[float]
+    """
+    Controls the loudness of the audio. Lower values result in quieter audio, 
+    while higher values make it louder. The suitable range is between 0.3 
+    and 3.0. Default is 1.0.
+    """
+
+    speech_sample_rate: typing_extensions.NotRequired[int]
+    """
+    Specifies the sample rate of the output audio. Supported values are 
+    8000, 16000, 22050, 24000 Hz. If not provided, the default is 22050 Hz.
+    """
+
+    enable_preprocessing: typing_extensions.NotRequired[bool]
+    """
+    Controls whether normalization of English words and numeric entities 
+    (e.g., numbers, dates) is performed. Set to true for better handling 
+    of mixed-language text. Default is false.
+    """
+
+    output_audio_codec: typing_extensions.NotRequired[typing.Literal["mp3"]]
+    """
+    Audio codec (currently supports MP3 only, optimized for real-time playback)
+    """
+
+    output_audio_bitrate: typing_extensions.NotRequired[InitializeConnectionDataOutputAudioBitrate]
+    """
+    Audio bitrate (choose from 5 supported bitrate options)
+    """
+
+    min_buffer_size: typing_extensions.NotRequired[int]
+    """
+    Minimum character length that triggers buffer flushing for TTS model processing
+    """
+
+    max_chunk_length: typing_extensions.NotRequired[int]
+    """
+    Maximum length for sentence splitting (adjust based on content length)
+    """

{sarvamai-0.1.8rc4 → sarvamai-0.1.8rc6}/src/sarvamai/requests/ping_signal.py

@@ -6,4 +6,9 @@ import typing_extensions
 
 
 class PingSignalParams(typing_extensions.TypedDict):
+    """
+    Send ping signal to keep the WebSocket connection alive. The connection automatically
+    closes after one minute of inactivity.
+    """
+
     type: typing.Literal["ping"]

sarvamai-0.1.8rc6/src/sarvamai/text_to_speech_streaming/__init__.py

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

{sarvamai-0.1.8rc4 → sarvamai-0.1.8rc6}/src/sarvamai/text_to_speech_streaming/client.py

@@ -11,7 +11,6 @@ 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:
@@ -33,7 +32,7 @@ class TextToSpeechStreamingClient:
     def connect(
         self,
         *,
-        model: typing.Optional[TextToSpeechStreamingModel] = None,
+        model: typing.Optional[typing.Literal["bulbul:v2"]] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[TextToSpeechStreamingSocketClient]:
@@ -43,7 +42,7 @@ class TextToSpeechStreamingClient:
 
         Parameters
         ----------
-        model : typing.Optional[TextToSpeechStreamingModel]
+        model : typing.Optional[typing.Literal["bulbul:v2"]]
             Text to speech model to use
 
         api_subscription_key : typing.Optional[str]
@@ -103,7 +102,7 @@ class AsyncTextToSpeechStreamingClient:
     async def connect(
         self,
         *,
-        model: typing.Optional[TextToSpeechStreamingModel] = None,
+        model: typing.Optional[typing.Literal["bulbul:v2"]] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncTextToSpeechStreamingSocketClient]:
@@ -113,7 +112,7 @@ class AsyncTextToSpeechStreamingClient:
 
         Parameters
         ----------
-        model : typing.Optional[TextToSpeechStreamingModel]
+        model : typing.Optional[typing.Literal["bulbul:v2"]]
             Text to speech model to use
 
         api_subscription_key : typing.Optional[str]

{sarvamai-0.1.8rc4 → sarvamai-0.1.8rc6}/src/sarvamai/text_to_speech_streaming/raw_client.py

@@ -10,7 +10,6 @@ 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:
@@ -21,7 +20,7 @@ class RawTextToSpeechStreamingClient:
     def connect(
         self,
         *,
-        model: typing.Optional[TextToSpeechStreamingModel] = None,
+        model: typing.Optional[typing.Literal["bulbul:v2"]] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[TextToSpeechStreamingSocketClient]:
@@ -31,7 +30,7 @@ class RawTextToSpeechStreamingClient:
 
         Parameters
         ----------
-        model : typing.Optional[TextToSpeechStreamingModel]
+        model : typing.Optional[typing.Literal["bulbul:v2"]]
             Text to speech model to use
 
         api_subscription_key : typing.Optional[str]
@@ -80,7 +79,7 @@ class AsyncRawTextToSpeechStreamingClient:
     async def connect(
         self,
         *,
-        model: typing.Optional[TextToSpeechStreamingModel] = None,
+        model: typing.Optional[typing.Literal["bulbul:v2"]] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncTextToSpeechStreamingSocketClient]:
@@ -90,7 +89,7 @@ class AsyncRawTextToSpeechStreamingClient:
 
         Parameters
         ----------
-        model : typing.Optional[TextToSpeechStreamingModel]
+        model : typing.Optional[typing.Literal["bulbul:v2"]]
             Text to speech model to use
 
         api_subscription_key : typing.Optional[str]

{sarvamai-0.1.8rc4 → sarvamai-0.1.8rc6}/src/sarvamai/text_to_speech_streaming/socket_client.py

@@ -8,7 +8,7 @@ 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_output import AudioOutput
-from ..types.close_connection import CloseConnection
+from ..types.flush_signal import FlushSignal
 from ..types.error_response import ErrorResponse
 from ..types.initialize_connection import InitializeConnection
 from ..types.initialize_connection_data import InitializeConnectionData
@@ -58,16 +58,43 @@ class AsyncTextToSpeechStreamingSocketClient(EventEmitterMixin):
         self,
         target_language_code: str,
         speaker: str,
+        pitch: float = 0.0,
+        pace: float = 1.0,
+        loudness: float = 1.0,
+        speech_sample_rate: int = 22050,
+        enable_preprocessing: bool = False,
+        output_audio_codec: str = "mp3",
+        output_audio_bitrate: str = "128k",
+        min_buffer_size: int = 50,
+        max_chunk_length: int = 150,
     ) -> None:
         """
         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')
+        :param pitch: Voice pitch adjustment (-1.0 to 1.0, default: 0.0)
+        :param pace: Speech pace (0.3 to 3.0, default: 1.0)
+        :param loudness: Voice loudness (0.1 to 3.0, default: 1.0)
+        :param speech_sample_rate: Audio sample rate, default: 22050
+        :param enable_preprocessing: Enable text preprocessing, default: False
+        :param output_audio_codec: Audio codec, default: 'mp3'
+        :param output_audio_bitrate: Audio bitrate, default: '128k'
+        :param min_buffer_size: Minimum buffer size, default: 50
+        :param max_chunk_length: Maximum chunk length, default: 150
         """
         data = InitializeConnectionData(
             target_language_code=target_language_code,
             speaker=speaker,
+            pitch=pitch,
+            pace=pace,
+            loudness=loudness,
+            speech_sample_rate=speech_sample_rate,
+            enable_preprocessing=enable_preprocessing,
+            output_audio_codec=output_audio_codec,
+            output_audio_bitrate=output_audio_bitrate,
+            min_buffer_size=min_buffer_size,
+            max_chunk_length=max_chunk_length,
         )
         message = InitializeConnection(data=data)
         await self._send_model(message)
@@ -87,7 +114,7 @@ class AsyncTextToSpeechStreamingSocketClient(EventEmitterMixin):
         Signal to flush the buffer and finalize audio output.
         This indicates the end of text input.
         """
-        message = CloseConnection()
+        message = FlushSignal()
         await self._send_model(message)
 
     async def ping(self) -> None:
@@ -159,16 +186,43 @@ class TextToSpeechStreamingSocketClient(EventEmitterMixin):
         self,
         target_language_code: str,
         speaker: str,
+        pitch: float = 0.0,
+        pace: float = 1.0,
+        loudness: float = 1.0,
+        speech_sample_rate: int = 22050,
+        enable_preprocessing: bool = False,
+        output_audio_codec: str = "mp3",
+        output_audio_bitrate: str = "128k",
+        min_buffer_size: int = 50,
+        max_chunk_length: int = 150,
     ) -> None:
         """
         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')
+        :param pitch: Voice pitch adjustment (-1.0 to 1.0, default: 0.0)
+        :param pace: Speech pace (0.3 to 3.0, default: 1.0)
+        :param loudness: Voice loudness (0.1 to 3.0, default: 1.0)
+        :param speech_sample_rate: Audio sample rate, default: 22050
+        :param enable_preprocessing: Enable text preprocessing, default: False
+        :param output_audio_codec: Audio codec, default: 'mp3'
+        :param output_audio_bitrate: Audio bitrate, default: '128k'
+        :param min_buffer_size: Minimum buffer size, default: 50
+        :param max_chunk_length: Maximum chunk length, default: 150
         """
         data = InitializeConnectionData(
             target_language_code=target_language_code,
             speaker=speaker,
+            pitch=pitch,
+            pace=pace,
+            loudness=loudness,
+            speech_sample_rate=speech_sample_rate,
+            enable_preprocessing=enable_preprocessing,
+            output_audio_codec=output_audio_codec,
+            output_audio_bitrate=output_audio_bitrate,
+            min_buffer_size=min_buffer_size,
+            max_chunk_length=max_chunk_length,
         )
         message = InitializeConnection(data=data)
         self._send_model(message)
@@ -188,7 +242,7 @@ class TextToSpeechStreamingSocketClient(EventEmitterMixin):
         Signal to flush the buffer and finalize audio output.
         This indicates the end of text input.
         """
-        message = CloseConnection()
+        message = FlushSignal()
         self._send_model(message)
 
     def ping(self) -> None:

{sarvamai-0.1.8rc4 → sarvamai-0.1.8rc6}/src/sarvamai/types/__init__.py

@@ -17,7 +17,6 @@ from .chat_completion_request_system_message import ChatCompletionRequestSystemM
 from .chat_completion_request_user_message import ChatCompletionRequestUserMessage
 from .chat_completion_response_message import ChatCompletionResponseMessage
 from .choice import Choice
-from .close_connection import CloseConnection
 from .completion_usage import CompletionUsage
 from .config_message import ConfigMessage
 from .create_chat_completion_response import CreateChatCompletionResponse
@@ -31,6 +30,7 @@ from .error_response import ErrorResponse
 from .error_response_data import ErrorResponseData
 from .events_data import EventsData
 from .finish_reason import FinishReason
+from .flush_signal import FlushSignal
 from .format import Format
 from .initialize_connection import InitializeConnection
 from .initialize_connection_data import InitializeConnectionData
@@ -92,7 +92,6 @@ __all__ = [
     "ChatCompletionRequestUserMessage",
     "ChatCompletionResponseMessage",
     "Choice",
-    "CloseConnection",
     "CompletionUsage",
     "ConfigMessage",
     "CreateChatCompletionResponse",
@@ -106,6 +105,7 @@ __all__ = [
     "ErrorResponseData",
     "EventsData",
     "FinishReason",
+    "FlushSignal",
     "Format",
     "InitializeConnection",
     "InitializeConnectionData",

sarvamai-0.1.8rc4/src/sarvamai/types/close_connection.py → sarvamai-0.1.8rc6/src/sarvamai/types/flush_signal.py

@@ -6,7 +6,12 @@ import pydantic
 from ..core.pydantic_utilities import IS_PYDANTIC_V2, UniversalBaseModel
 
 
-class CloseConnection(UniversalBaseModel):
+class FlushSignal(UniversalBaseModel):
+    """
+    Forces the text buffer to process immediately, regardless of the min_buffer_size threshold.
+    Use this when you need to process remaining text that hasn't reached the minimum buffer size.
+    """
+
     type: typing.Literal["flush"] = "flush"
 
     if IS_PYDANTIC_V2:

{sarvamai-0.1.8rc4 → sarvamai-0.1.8rc6}/src/sarvamai/types/initialize_connection.py

@@ -8,6 +8,13 @@ from .initialize_connection_data import InitializeConnectionData
 
 
 class InitializeConnection(UniversalBaseModel):
+    """
+    Configuration message required as the first message after establishing the WebSocket connection.
+    This initializes TTS parameters and can be updated at any time during the WebSocket lifecycle
+    by sending a new config message. When a config update is sent, any text currently in the buffer
+    will be automatically flushed and processed before applying the new configuration.
+    """
+
     type: typing.Literal["config"] = "config"
     data: InitializeConnectionData
 

sarvamai-0.1.8rc6/src/sarvamai/types/initialize_connection_data.py

@@ -0,0 +1,93 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+import pydantic
+from ..core.pydantic_utilities import IS_PYDANTIC_V2, UniversalBaseModel
+from .initialize_connection_data_output_audio_bitrate import InitializeConnectionDataOutputAudioBitrate
+from .initialize_connection_data_speaker import InitializeConnectionDataSpeaker
+from .initialize_connection_data_target_language_code import InitializeConnectionDataTargetLanguageCode
+
+
+class InitializeConnectionData(UniversalBaseModel):
+    target_language_code: InitializeConnectionDataTargetLanguageCode = pydantic.Field()
+    """
+    The language of the text is BCP-47 format
+    """
+
+    speaker: InitializeConnectionDataSpeaker = pydantic.Field()
+    """
+    The speaker voice to be used for the output audio.
+    
+    **Default:** Anushka
+    
+    **Model Compatibility (Speakers compatible with respective model):**
+    - **bulbul:v2:**
+      - Female: Anushka, Manisha, Vidya, Arya
+      - Male: Abhilash, Karun, Hitesh
+    
+    **Note:** Speaker selection must match the chosen model version.
+    """
+
+    pitch: typing.Optional[float] = pydantic.Field(default=None)
+    """
+    Controls the pitch of the audio. Lower values result in a deeper voice, 
+    while higher values make it sharper. The suitable range is between -0.75 
+    and 0.75. Default is 0.0.
+    """
+
+    pace: typing.Optional[float] = pydantic.Field(default=None)
+    """
+    Controls the speed of the audio. Lower values result in slower speech, 
+    while higher values make it faster. The suitable range is between 0.5 
+    and 2.0. Default is 1.0.
+    """
+
+    loudness: typing.Optional[float] = pydantic.Field(default=None)
+    """
+    Controls the loudness of the audio. Lower values result in quieter audio, 
+    while higher values make it louder. The suitable range is between 0.3 
+    and 3.0. Default is 1.0.
+    """
+
+    speech_sample_rate: typing.Optional[int] = pydantic.Field(default=None)
+    """
+    Specifies the sample rate of the output audio. Supported values are 
+    8000, 16000, 22050, 24000 Hz. If not provided, the default is 22050 Hz.
+    """
+
+    enable_preprocessing: typing.Optional[bool] = pydantic.Field(default=None)
+    """
+    Controls whether normalization of English words and numeric entities 
+    (e.g., numbers, dates) is performed. Set to true for better handling 
+    of mixed-language text. Default is false.
+    """
+
+    output_audio_codec: typing.Optional[typing.Literal["mp3"]] = pydantic.Field(default=None)
+    """
+    Audio codec (currently supports MP3 only, optimized for real-time playback)
+    """
+
+    output_audio_bitrate: typing.Optional[InitializeConnectionDataOutputAudioBitrate] = pydantic.Field(default=None)
+    """
+    Audio bitrate (choose from 5 supported bitrate options)
+    """
+
+    min_buffer_size: typing.Optional[int] = pydantic.Field(default=None)
+    """
+    Minimum character length that triggers buffer flushing for TTS model processing
+    """
+
+    max_chunk_length: typing.Optional[int] = pydantic.Field(default=None)
+    """
+    Maximum length for sentence splitting (adjust based on content length)
+    """
+
+    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-0.1.8rc6/src/sarvamai/types/initialize_connection_data_speaker.py

@@ -0,0 +1,7 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+InitializeConnectionDataSpeaker = typing.Union[
+    typing.Literal["anushka", "abhilash", "manisha", "vidya", "arya", "karun", "hitesh"], typing.Any
+]

{sarvamai-0.1.8rc4 → sarvamai-0.1.8rc6}/src/sarvamai/types/ping_signal.py

@@ -7,6 +7,11 @@ from ..core.pydantic_utilities import IS_PYDANTIC_V2, UniversalBaseModel
 
 
 class PingSignal(UniversalBaseModel):
+    """
+    Send ping signal to keep the WebSocket connection alive. The connection automatically
+    closes after one minute of inactivity.
+    """
+
     type: typing.Literal["ping"] = "ping"
 
     if IS_PYDANTIC_V2:

sarvamai-0.1.8rc4/src/sarvamai/requests/close_connection.py

@@ -1,9 +0,0 @@
-# This file was auto-generated by Fern from our API Definition.
-
-import typing
-
-import typing_extensions
-
-
-class CloseConnectionParams(typing_extensions.TypedDict):
-    type: typing.Literal["flush"]

sarvamai-0.1.8rc4/src/sarvamai/requests/initialize_connection.py

@@ -1,11 +0,0 @@
-# This file was auto-generated by Fern from our API Definition.
-
-import typing
-
-import typing_extensions
-from .initialize_connection_data import InitializeConnectionDataParams
-
-
-class InitializeConnectionParams(typing_extensions.TypedDict):
-    type: typing.Literal["config"]
-    data: InitializeConnectionDataParams