sarvamai 0.1.23a3__py3-none-any.whl → 0.1.23a4__py3-none-any.whl

sarvamai/text_to_speech_streaming/client.py

@@ -11,6 +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
 from .types.text_to_speech_streaming_send_completion_event import TextToSpeechStreamingSendCompletionEvent
 
 try:
@@ -38,7 +39,7 @@ class TextToSpeechStreamingClient:
     def connect(
         self,
         *,
-        model: typing.Optional[typing.Literal["bulbul:v2"]] = None,
+        model: typing.Optional[TextToSpeechStreamingModel] = None,
         send_completion_event: typing.Optional[TextToSpeechStreamingSendCompletionEvent] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
@@ -50,10 +51,16 @@ class TextToSpeechStreamingClient:
         The Try It playground may not provide the best experience for streaming audio.
         For optimal streaming performance, please use the SDK or implement your own WebSocket client.
 
+        **Model-Specific Notes:**
+        - **bulbul:v2:** Supports pitch, loudness, pace (0.3-3.0). Default sample rate: 22050 Hz.
+        - **bulbul:v3-beta:** Does NOT support pitch/loudness. Pace range: 0.5-2.0. Supports temperature parameter. Default sample rate: 24000 Hz. Preprocessing is always enabled.
+
         Parameters
         ----------
-        model : typing.Optional[typing.Literal["bulbul:v2"]]
-            Text to speech model to use
+        model : typing.Optional[TextToSpeechStreamingModel]
+            Text to speech model to use.
+            - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
+            - **bulbul:v3-beta**: Advanced model with temperature control (no pitch/loudness)
 
         send_completion_event : typing.Optional[TextToSpeechStreamingSendCompletionEvent]
             Enable completion event notifications when TTS generation finishes. When set to true, an event message will be sent when the final audio chunk has been generated.
@@ -117,7 +124,7 @@ class AsyncTextToSpeechStreamingClient:
     async def connect(
         self,
         *,
-        model: typing.Optional[typing.Literal["bulbul:v2"]] = None,
+        model: typing.Optional[TextToSpeechStreamingModel] = None,
         send_completion_event: typing.Optional[TextToSpeechStreamingSendCompletionEvent] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
@@ -129,10 +136,16 @@ class AsyncTextToSpeechStreamingClient:
         The Try It playground may not provide the best experience for streaming audio.
         For optimal streaming performance, please use the SDK or implement your own WebSocket client.
 
+        **Model-Specific Notes:**
+        - **bulbul:v2:** Supports pitch, loudness, pace (0.3-3.0). Default sample rate: 22050 Hz.
+        - **bulbul:v3-beta:** Does NOT support pitch/loudness. Pace range: 0.5-2.0. Supports temperature parameter. Default sample rate: 24000 Hz. Preprocessing is always enabled.
+
         Parameters
         ----------
-        model : typing.Optional[typing.Literal["bulbul:v2"]]
-            Text to speech model to use
+        model : typing.Optional[TextToSpeechStreamingModel]
+            Text to speech model to use.
+            - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
+            - **bulbul:v3-beta**: Advanced model with temperature control (no pitch/loudness)
 
         send_completion_event : typing.Optional[TextToSpeechStreamingSendCompletionEvent]
             Enable completion event notifications when TTS generation finishes. When set to true, an event message will be sent when the final audio chunk has been generated.

sarvamai/text_to_speech_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 AsyncTextToSpeechStreamingSocketClient, TextToSpeechStreamingSocketClient
+from .types.text_to_speech_streaming_model import TextToSpeechStreamingModel
 from .types.text_to_speech_streaming_send_completion_event import TextToSpeechStreamingSendCompletionEvent
 
 try:
@@ -26,7 +27,7 @@ class RawTextToSpeechStreamingClient:
     def connect(
         self,
         *,
-        model: typing.Optional[typing.Literal["bulbul:v2"]] = None,
+        model: typing.Optional[TextToSpeechStreamingModel] = None,
         send_completion_event: typing.Optional[TextToSpeechStreamingSendCompletionEvent] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
@@ -38,10 +39,16 @@ class RawTextToSpeechStreamingClient:
         The Try It playground may not provide the best experience for streaming audio.
         For optimal streaming performance, please use the SDK or implement your own WebSocket client.
 
+        **Model-Specific Notes:**
+        - **bulbul:v2:** Supports pitch, loudness, pace (0.3-3.0). Default sample rate: 22050 Hz.
+        - **bulbul:v3-beta:** Does NOT support pitch/loudness. Pace range: 0.5-2.0. Supports temperature parameter. Default sample rate: 24000 Hz. Preprocessing is always enabled.
+
         Parameters
         ----------
-        model : typing.Optional[typing.Literal["bulbul:v2"]]
-            Text to speech model to use
+        model : typing.Optional[TextToSpeechStreamingModel]
+            Text to speech model to use.
+            - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
+            - **bulbul:v3-beta**: Advanced model with temperature control (no pitch/loudness)
 
         send_completion_event : typing.Optional[TextToSpeechStreamingSendCompletionEvent]
             Enable completion event notifications when TTS generation finishes. When set to true, an event message will be sent when the final audio chunk has been generated.
@@ -94,7 +101,7 @@ class AsyncRawTextToSpeechStreamingClient:
     async def connect(
         self,
         *,
-        model: typing.Optional[typing.Literal["bulbul:v2"]] = None,
+        model: typing.Optional[TextToSpeechStreamingModel] = None,
         send_completion_event: typing.Optional[TextToSpeechStreamingSendCompletionEvent] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
@@ -106,10 +113,16 @@ class AsyncRawTextToSpeechStreamingClient:
         The Try It playground may not provide the best experience for streaming audio.
         For optimal streaming performance, please use the SDK or implement your own WebSocket client.
 
+        **Model-Specific Notes:**
+        - **bulbul:v2:** Supports pitch, loudness, pace (0.3-3.0). Default sample rate: 22050 Hz.
+        - **bulbul:v3-beta:** Does NOT support pitch/loudness. Pace range: 0.5-2.0. Supports temperature parameter. Default sample rate: 24000 Hz. Preprocessing is always enabled.
+
         Parameters
         ----------
-        model : typing.Optional[typing.Literal["bulbul:v2"]]
-            Text to speech model to use
+        model : typing.Optional[TextToSpeechStreamingModel]
+            Text to speech model to use.
+            - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
+            - **bulbul:v3-beta**: Advanced model with temperature control (no pitch/loudness)
 
         send_completion_event : typing.Optional[TextToSpeechStreamingSendCompletionEvent]
             Enable completion event notifications when TTS generation finishes. When set to true, an event message will be sent when the final audio chunk has been generated.

sarvamai/text_to_speech_streaming/types/__init__.py

@@ -2,35 +2,7 @@
 
 # isort: skip_file
 
-import typing
-from importlib import import_module
+from .text_to_speech_streaming_model import TextToSpeechStreamingModel
+from .text_to_speech_streaming_send_completion_event import TextToSpeechStreamingSendCompletionEvent
 
-if typing.TYPE_CHECKING:
-    from .text_to_speech_streaming_send_completion_event import TextToSpeechStreamingSendCompletionEvent
-_dynamic_imports: typing.Dict[str, str] = {
-    "TextToSpeechStreamingSendCompletionEvent": ".text_to_speech_streaming_send_completion_event"
-}
-
-
-def __getattr__(attr_name: str) -> typing.Any:
-    module_name = _dynamic_imports.get(attr_name)
-    if module_name is None:
-        raise AttributeError(f"No {attr_name} found in _dynamic_imports for module name -> {__name__}")
-    try:
-        module = import_module(module_name, __package__)
-        if module_name == f".{attr_name}":
-            return module
-        else:
-            return getattr(module, attr_name)
-    except ImportError as e:
-        raise ImportError(f"Failed to import {attr_name} from {module_name}: {e}") from e
-    except AttributeError as e:
-        raise AttributeError(f"Failed to get {attr_name} from {module_name}: {e}") from e
-
-
-def __dir__():
-    lazy_attrs = list(_dynamic_imports.keys())
-    return sorted(lazy_attrs)
-
-
-__all__ = ["TextToSpeechStreamingSendCompletionEvent"]
+__all__ = ["TextToSpeechStreamingModel", "TextToSpeechStreamingSendCompletionEvent"]

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:v2", "bulbul:v3-beta"], typing.Any]

sarvamai/types/__init__.py

@@ -2,228 +2,106 @@
 
 # isort: skip_file
 
-import typing
-from importlib import import_module
-
-if typing.TYPE_CHECKING:
-    from .audio_data import AudioData
-    from .audio_message import AudioMessage
-    from .audio_output import AudioOutput
-    from .audio_output_data import AudioOutputData
-    from .base_job_parameters import BaseJobParameters
-    from .bulk_job_callback import BulkJobCallback
-    from .bulk_job_init_response_v_1 import BulkJobInitResponseV1
-    from .chat_completion_request_assistant_message import ChatCompletionRequestAssistantMessage
-    from .chat_completion_request_message import (
-        ChatCompletionRequestMessage,
-        ChatCompletionRequestMessage_Assistant,
-        ChatCompletionRequestMessage_System,
-        ChatCompletionRequestMessage_User,
-    )
-    from .chat_completion_request_system_message import ChatCompletionRequestSystemMessage
-    from .chat_completion_request_user_message import ChatCompletionRequestUserMessage
-    from .chat_completion_response_message import ChatCompletionResponseMessage
-    from .choice import Choice
-    from .completion_event_flag import CompletionEventFlag
-    from .completion_usage import CompletionUsage
-    from .config_message import ConfigMessage
-    from .configure_connection import ConfigureConnection
-    from .configure_connection_data import ConfigureConnectionData
-    from .configure_connection_data_output_audio_bitrate import ConfigureConnectionDataOutputAudioBitrate
-    from .configure_connection_data_output_audio_codec import ConfigureConnectionDataOutputAudioCodec
-    from .configure_connection_data_speaker import ConfigureConnectionDataSpeaker
-    from .configure_connection_data_target_language_code import ConfigureConnectionDataTargetLanguageCode
-    from .connection_sample_rate import ConnectionSampleRate
-    from .create_chat_completion_response import CreateChatCompletionResponse
-    from .diarized_entry import DiarizedEntry
-    from .diarized_transcript import DiarizedTranscript
-    from .error_code import ErrorCode
-    from .error_data import ErrorData
-    from .error_details import ErrorDetails
-    from .error_message import ErrorMessage
-    from .error_response import ErrorResponse
-    from .error_response_data import ErrorResponseData
-    from .event_response import EventResponse
-    from .event_response_data import EventResponseData
-    from .events_data import EventsData
-    from .events_data_signal_type import EventsDataSignalType
-    from .file_signed_url_details import FileSignedUrlDetails
-    from .files_download_response import FilesDownloadResponse
-    from .files_request import FilesRequest
-    from .files_upload_response import FilesUploadResponse
-    from .finish_reason import FinishReason
-    from .flush_signal import FlushSignal
-    from .input_audio_codec import InputAudioCodec
-    from .job_state import JobState
-    from .job_status_v_1_response import JobStatusV1Response
-    from .language_identification_response import LanguageIdentificationResponse
-    from .numerals_format import NumeralsFormat
-    from .ping_signal import PingSignal
-    from .reasoning_effort import ReasoningEffort
-    from .response_type import ResponseType
-    from .role import Role
-    from .sarvam_model_ids import SarvamModelIds
-    from .send_text import SendText
-    from .send_text_data import SendTextData
-    from .speech_sample_rate import SpeechSampleRate
-    from .speech_to_text_job_parameters import SpeechToTextJobParameters
-    from .speech_to_text_language import SpeechToTextLanguage
-    from .speech_to_text_model import SpeechToTextModel
-    from .speech_to_text_response import SpeechToTextResponse
-    from .speech_to_text_response_data import SpeechToTextResponseData
-    from .speech_to_text_streaming_response import SpeechToTextStreamingResponse
-    from .speech_to_text_transcription_data import SpeechToTextTranscriptionData
-    from .speech_to_text_translate_job_parameters import SpeechToTextTranslateJobParameters
-    from .speech_to_text_translate_language import SpeechToTextTranslateLanguage
-    from .speech_to_text_translate_model import SpeechToTextTranslateModel
-    from .speech_to_text_translate_response import SpeechToTextTranslateResponse
-    from .speech_to_text_translate_response_data import SpeechToTextTranslateResponseData
-    from .speech_to_text_translate_streaming_response import SpeechToTextTranslateStreamingResponse
-    from .speech_to_text_translate_transcription_data import SpeechToTextTranslateTranscriptionData
-    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
-    from .text_to_speech_language import TextToSpeechLanguage
-    from .text_to_speech_model import TextToSpeechModel
-    from .text_to_speech_output_audio_codec import TextToSpeechOutputAudioCodec
-    from .text_to_speech_response import TextToSpeechResponse
-    from .text_to_speech_speaker import TextToSpeechSpeaker
-    from .timestamps_model import TimestampsModel
-    from .transcription_metrics import TranscriptionMetrics
-    from .translate_mode import TranslateMode
-    from .translate_model import TranslateModel
-    from .translate_source_language import TranslateSourceLanguage
-    from .translate_speaker_gender import TranslateSpeakerGender
-    from .translate_target_language import TranslateTargetLanguage
-    from .translation_response import TranslationResponse
-    from .translatiterate_target_language import TranslatiterateTargetLanguage
-    from .transliterate_mode import TransliterateMode
-    from .transliterate_source_language import TransliterateSourceLanguage
-    from .transliteration_response import TransliterationResponse
-_dynamic_imports: typing.Dict[str, str] = {
-    "AudioData": ".audio_data",
-    "AudioMessage": ".audio_message",
-    "AudioOutput": ".audio_output",
-    "AudioOutputData": ".audio_output_data",
-    "BaseJobParameters": ".base_job_parameters",
-    "BulkJobCallback": ".bulk_job_callback",
-    "BulkJobInitResponseV1": ".bulk_job_init_response_v_1",
-    "ChatCompletionRequestAssistantMessage": ".chat_completion_request_assistant_message",
-    "ChatCompletionRequestMessage": ".chat_completion_request_message",
-    "ChatCompletionRequestMessage_Assistant": ".chat_completion_request_message",
-    "ChatCompletionRequestMessage_System": ".chat_completion_request_message",
-    "ChatCompletionRequestMessage_User": ".chat_completion_request_message",
-    "ChatCompletionRequestSystemMessage": ".chat_completion_request_system_message",
-    "ChatCompletionRequestUserMessage": ".chat_completion_request_user_message",
-    "ChatCompletionResponseMessage": ".chat_completion_response_message",
-    "Choice": ".choice",
-    "CompletionEventFlag": ".completion_event_flag",
-    "CompletionUsage": ".completion_usage",
-    "ConfigMessage": ".config_message",
-    "ConfigureConnection": ".configure_connection",
-    "ConfigureConnectionData": ".configure_connection_data",
-    "ConfigureConnectionDataOutputAudioBitrate": ".configure_connection_data_output_audio_bitrate",
-    "ConfigureConnectionDataOutputAudioCodec": ".configure_connection_data_output_audio_codec",
-    "ConfigureConnectionDataSpeaker": ".configure_connection_data_speaker",
-    "ConfigureConnectionDataTargetLanguageCode": ".configure_connection_data_target_language_code",
-    "ConnectionSampleRate": ".connection_sample_rate",
-    "CreateChatCompletionResponse": ".create_chat_completion_response",
-    "DiarizedEntry": ".diarized_entry",
-    "DiarizedTranscript": ".diarized_transcript",
-    "ErrorCode": ".error_code",
-    "ErrorData": ".error_data",
-    "ErrorDetails": ".error_details",
-    "ErrorMessage": ".error_message",
-    "ErrorResponse": ".error_response",
-    "ErrorResponseData": ".error_response_data",
-    "EventResponse": ".event_response",
-    "EventResponseData": ".event_response_data",
-    "EventsData": ".events_data",
-    "EventsDataSignalType": ".events_data_signal_type",
-    "FileSignedUrlDetails": ".file_signed_url_details",
-    "FilesDownloadResponse": ".files_download_response",
-    "FilesRequest": ".files_request",
-    "FilesUploadResponse": ".files_upload_response",
-    "FinishReason": ".finish_reason",
-    "FlushSignal": ".flush_signal",
-    "InputAudioCodec": ".input_audio_codec",
-    "JobState": ".job_state",
-    "JobStatusV1Response": ".job_status_v_1_response",
-    "LanguageIdentificationResponse": ".language_identification_response",
-    "NumeralsFormat": ".numerals_format",
-    "PingSignal": ".ping_signal",
-    "ReasoningEffort": ".reasoning_effort",
-    "ResponseType": ".response_type",
-    "Role": ".role",
-    "SarvamModelIds": ".sarvam_model_ids",
-    "SendText": ".send_text",
-    "SendTextData": ".send_text_data",
-    "SpeechSampleRate": ".speech_sample_rate",
-    "SpeechToTextJobParameters": ".speech_to_text_job_parameters",
-    "SpeechToTextLanguage": ".speech_to_text_language",
-    "SpeechToTextModel": ".speech_to_text_model",
-    "SpeechToTextResponse": ".speech_to_text_response",
-    "SpeechToTextResponseData": ".speech_to_text_response_data",
-    "SpeechToTextStreamingResponse": ".speech_to_text_streaming_response",
-    "SpeechToTextTranscriptionData": ".speech_to_text_transcription_data",
-    "SpeechToTextTranslateJobParameters": ".speech_to_text_translate_job_parameters",
-    "SpeechToTextTranslateLanguage": ".speech_to_text_translate_language",
-    "SpeechToTextTranslateModel": ".speech_to_text_translate_model",
-    "SpeechToTextTranslateResponse": ".speech_to_text_translate_response",
-    "SpeechToTextTranslateResponseData": ".speech_to_text_translate_response_data",
-    "SpeechToTextTranslateStreamingResponse": ".speech_to_text_translate_streaming_response",
-    "SpeechToTextTranslateTranscriptionData": ".speech_to_text_translate_transcription_data",
-    "SpokenFormNumeralsFormat": ".spoken_form_numerals_format",
-    "StopConfiguration": ".stop_configuration",
-    "StorageContainerType": ".storage_container_type",
-    "SttFlushSignal": ".stt_flush_signal",
-    "TaskDetailV1": ".task_detail_v_1",
-    "TaskFileDetails": ".task_file_details",
-    "TaskState": ".task_state",
-    "TextToSpeechLanguage": ".text_to_speech_language",
-    "TextToSpeechModel": ".text_to_speech_model",
-    "TextToSpeechOutputAudioCodec": ".text_to_speech_output_audio_codec",
-    "TextToSpeechResponse": ".text_to_speech_response",
-    "TextToSpeechSpeaker": ".text_to_speech_speaker",
-    "TimestampsModel": ".timestamps_model",
-    "TranscriptionMetrics": ".transcription_metrics",
-    "TranslateMode": ".translate_mode",
-    "TranslateModel": ".translate_model",
-    "TranslateSourceLanguage": ".translate_source_language",
-    "TranslateSpeakerGender": ".translate_speaker_gender",
-    "TranslateTargetLanguage": ".translate_target_language",
-    "TranslationResponse": ".translation_response",
-    "TranslatiterateTargetLanguage": ".translatiterate_target_language",
-    "TransliterateMode": ".transliterate_mode",
-    "TransliterateSourceLanguage": ".transliterate_source_language",
-    "TransliterationResponse": ".transliteration_response",
-}
-
-
-def __getattr__(attr_name: str) -> typing.Any:
-    module_name = _dynamic_imports.get(attr_name)
-    if module_name is None:
-        raise AttributeError(f"No {attr_name} found in _dynamic_imports for module name -> {__name__}")
-    try:
-        module = import_module(module_name, __package__)
-        if module_name == f".{attr_name}":
-            return module
-        else:
-            return getattr(module, attr_name)
-    except ImportError as e:
-        raise ImportError(f"Failed to import {attr_name} from {module_name}: {e}") from e
-    except AttributeError as e:
-        raise AttributeError(f"Failed to get {attr_name} from {module_name}: {e}") from e
-
-
-def __dir__():
-    lazy_attrs = list(_dynamic_imports.keys())
-    return sorted(lazy_attrs)
-
+from .audio_data import AudioData
+from .audio_message import AudioMessage
+from .audio_output import AudioOutput
+from .audio_output_data import AudioOutputData
+from .base_job_parameters import BaseJobParameters
+from .bulk_job_callback import BulkJobCallback
+from .bulk_job_init_response_v_1 import BulkJobInitResponseV1
+from .chat_completion_request_assistant_message import ChatCompletionRequestAssistantMessage
+from .chat_completion_request_message import (
+    ChatCompletionRequestMessage,
+    ChatCompletionRequestMessage_Assistant,
+    ChatCompletionRequestMessage_System,
+    ChatCompletionRequestMessage_User,
+)
+from .chat_completion_request_system_message import ChatCompletionRequestSystemMessage
+from .chat_completion_request_user_message import ChatCompletionRequestUserMessage
+from .chat_completion_response_message import ChatCompletionResponseMessage
+from .choice import Choice
+from .completion_event_flag import CompletionEventFlag
+from .completion_usage import CompletionUsage
+from .config_message import ConfigMessage
+from .configure_connection import ConfigureConnection
+from .configure_connection_data import ConfigureConnectionData
+from .configure_connection_data_model import ConfigureConnectionDataModel
+from .configure_connection_data_output_audio_bitrate import ConfigureConnectionDataOutputAudioBitrate
+from .configure_connection_data_output_audio_codec import ConfigureConnectionDataOutputAudioCodec
+from .configure_connection_data_speaker import ConfigureConnectionDataSpeaker
+from .configure_connection_data_target_language_code import ConfigureConnectionDataTargetLanguageCode
+from .connection_sample_rate import ConnectionSampleRate
+from .create_chat_completion_response import CreateChatCompletionResponse
+from .diarized_entry import DiarizedEntry
+from .diarized_transcript import DiarizedTranscript
+from .error_code import ErrorCode
+from .error_data import ErrorData
+from .error_details import ErrorDetails
+from .error_message import ErrorMessage
+from .error_response import ErrorResponse
+from .error_response_data import ErrorResponseData
+from .event_response import EventResponse
+from .event_response_data import EventResponseData
+from .events_data import EventsData
+from .events_data_signal_type import EventsDataSignalType
+from .file_signed_url_details import FileSignedUrlDetails
+from .files_download_response import FilesDownloadResponse
+from .files_request import FilesRequest
+from .files_upload_response import FilesUploadResponse
+from .finish_reason import FinishReason
+from .flush_signal import FlushSignal
+from .input_audio_codec import InputAudioCodec
+from .job_state import JobState
+from .job_status_v_1_response import JobStatusV1Response
+from .language_identification_response import LanguageIdentificationResponse
+from .mode import Mode
+from .numerals_format import NumeralsFormat
+from .ping_signal import PingSignal
+from .reasoning_effort import ReasoningEffort
+from .response_type import ResponseType
+from .role import Role
+from .sarvam_model_ids import SarvamModelIds
+from .send_text import SendText
+from .send_text_data import SendTextData
+from .speech_sample_rate import SpeechSampleRate
+from .speech_to_text_job_parameters import SpeechToTextJobParameters
+from .speech_to_text_language import SpeechToTextLanguage
+from .speech_to_text_model import SpeechToTextModel
+from .speech_to_text_response import SpeechToTextResponse
+from .speech_to_text_response_data import SpeechToTextResponseData
+from .speech_to_text_streaming_response import SpeechToTextStreamingResponse
+from .speech_to_text_transcription_data import SpeechToTextTranscriptionData
+from .speech_to_text_translate_job_parameters import SpeechToTextTranslateJobParameters
+from .speech_to_text_translate_language import SpeechToTextTranslateLanguage
+from .speech_to_text_translate_model import SpeechToTextTranslateModel
+from .speech_to_text_translate_response import SpeechToTextTranslateResponse
+from .speech_to_text_translate_response_data import SpeechToTextTranslateResponseData
+from .speech_to_text_translate_streaming_response import SpeechToTextTranslateStreamingResponse
+from .speech_to_text_translate_transcription_data import SpeechToTextTranslateTranscriptionData
+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
+from .text_to_speech_language import TextToSpeechLanguage
+from .text_to_speech_model import TextToSpeechModel
+from .text_to_speech_output_audio_codec import TextToSpeechOutputAudioCodec
+from .text_to_speech_response import TextToSpeechResponse
+from .text_to_speech_speaker import TextToSpeechSpeaker
+from .timestamps_model import TimestampsModel
+from .transcription_metrics import TranscriptionMetrics
+from .translate_mode import TranslateMode
+from .translate_model import TranslateModel
+from .translate_source_language import TranslateSourceLanguage
+from .translate_speaker_gender import TranslateSpeakerGender
+from .translate_target_language import TranslateTargetLanguage
+from .translation_response import TranslationResponse
+from .translatiterate_target_language import TranslatiterateTargetLanguage
+from .transliterate_mode import TransliterateMode
+from .transliterate_source_language import TransliterateSourceLanguage
+from .transliteration_response import TransliterationResponse
 
 __all__ = [
     "AudioData",
@@ -247,6 +125,7 @@ __all__ = [
     "ConfigMessage",
     "ConfigureConnection",
     "ConfigureConnectionData",
+    "ConfigureConnectionDataModel",
     "ConfigureConnectionDataOutputAudioBitrate",
     "ConfigureConnectionDataOutputAudioCodec",
     "ConfigureConnectionDataSpeaker",
@@ -275,6 +154,7 @@ __all__ = [
     "JobState",
     "JobStatusV1Response",
     "LanguageIdentificationResponse",
+    "Mode",
     "NumeralsFormat",
     "PingSignal",
     "ReasoningEffort",

sarvamai/types/chat_completion_request_message.py

@@ -5,7 +5,6 @@ from __future__ import annotations
 import typing
 
 import pydantic
-import typing_extensions
 from ..core.pydantic_utilities import IS_PYDANTIC_V2, UniversalBaseModel
 
 
@@ -51,9 +50,6 @@ class ChatCompletionRequestMessage_User(UniversalBaseModel):
             extra = pydantic.Extra.allow
 
 
-ChatCompletionRequestMessage = typing_extensions.Annotated[
-    typing.Union[
-        ChatCompletionRequestMessage_Assistant, ChatCompletionRequestMessage_System, ChatCompletionRequestMessage_User
-    ],
-    pydantic.Field(discriminator="role"),
+ChatCompletionRequestMessage = typing.Union[
+    ChatCompletionRequestMessage_Assistant, ChatCompletionRequestMessage_System, ChatCompletionRequestMessage_User
 ]

sarvamai/types/configure_connection.py

@@ -13,6 +13,10 @@ class ConfigureConnection(UniversalBaseModel):
     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.
+
+    **Model-Specific Notes:**
+    - **bulbul:v2:** Supports pitch, loudness, pace (0.3-3.0). Default sample rate: 22050 Hz.
+    - **bulbul:v3-beta:** Does NOT support pitch/loudness. Pace range: 0.5-2.0. Supports temperature. Default sample rate: 24000 Hz.
     """
 
     type: typing.Literal["config"] = "config"