sarvamai 0.1.22a4__py3-none-any.whl → 0.1.22a7__py3-none-any.whl

sarvamai/text_to_speech/raw_client.py

@@ -41,15 +41,38 @@ class RawTextToSpeechClient:
         enable_preprocessing: typing.Optional[bool] = OMIT,
         model: typing.Optional[TextToSpeechModel] = OMIT,
         output_audio_codec: typing.Optional[TextToSpeechOutputAudioCodec] = OMIT,
+        temperature: typing.Optional[float] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> HttpResponse[TextToSpeechResponse]:
         """
-        This is the model to convert text into spoken audio.
-        The output is a wave file encoded as a base64 string.
+        Convert text into spoken audio. The output is a wave file encoded as a base64 string.
+
+        **Available Models:**
+        - **bulbul:v2** (default): Supports pitch, loudness, and pace controls
+        - **bulbul:v3-beta**: Newer model with temperature control and improved quality
+
+        **Important Notes for bulbul:v3-beta:**
+        - Pitch and loudness parameters are NOT supported
+        - Pace must be between 0.5 and 2.0
+        - Preprocessing is automatically enabled
+        - Default sample rate is 24000 Hz
+        - Temperature parameter available (0.01-1.0, default 0.6)
 
         Parameters
         ----------
         text : str
+            The text(s) to be converted into speech.
+
+            **Features:**
+            - Supports code-mixed text (English and Indic languages)
+
+            **Model-specific limits:**
+            - **bulbul:v2:** Max 1500 characters
+            - **bulbul:v3-beta:** Max 2500 characters
+
+            **Important Note:**
+            - For numbers larger than 4 digits, use commas (e.g., '10,000' instead of '10000')
+            - This ensures proper pronunciation as a whole number
 
         target_language_code : TextToSpeechLanguage
             The language of the text is BCP-47 format
@@ -57,36 +80,63 @@ class RawTextToSpeechClient:
         speaker : typing.Optional[TextToSpeechSpeaker]
             The speaker voice to be used for the output audio.
 
-            **Default:** Anushka
+            **Default:** Anushka (for bulbul:v2), Aditya (for bulbul:v3-beta)
 
             **Model Compatibility (Speakers compatible with respective model):**
             - **bulbul:v2:**
               - Female: Anushka, Manisha, Vidya, Arya
               - Male: Abhilash, Karun, Hitesh
+            - **bulbul:v3-beta:**
+              - Aditya, Ritu, Priya, Neha, Rahul, Pooja, Rohan, Simran, Kavya, Amit, Dev, Ishita, Shreya, Ratan, Varun, Manan, Sumit, Roopa, Kabir, Aayan, Shubh, Ashutosh, Advait, Amelia, Sophia
 
             **Note:** Speaker selection must match the chosen model version.
 
         pitch : typing.Optional[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.
 
+            **Note:** This parameter is only supported for bulbul:v2. It is NOT supported for bulbul:v3-beta and will cause a validation error if provided.
+
         pace : typing.Optional[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.
+            Controls the speed of the audio. Lower values result in slower speech, while higher values make it faster. Default is 1.0.
+
+            **Model-specific ranges:**
+            - **bulbul:v2:** 0.3 to 3.0
+            - **bulbul:v3-beta:** 0.5 to 2.0
 
         loudness : typing.Optional[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.
 
+            **Note:** This parameter is only supported for bulbul:v2. It is NOT supported for bulbul:v3-beta and will cause a validation error if provided.
+
         speech_sample_rate : typing.Optional[SpeechSampleRate]
-            Specifies the sample rate of the output audio. Supported values are 8000, 16000, 22050, 24000 Hz. If not provided, the default is 22050 Hz.
+            Specifies the sample rate of the output audio. Supported values are 8000, 16000, 22050, 24000 Hz.
+
+            **Model-specific defaults:**
+            - **bulbul:v2:** Default is 22050 Hz
+            - **bulbul:v3-beta:** Default is 24000 Hz
 
         enable_preprocessing : typing.Optional[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.
+            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.
+
+            **Model-specific behavior:**
+            - **bulbul:v2:** Default is false
+            - **bulbul:v3-beta:** Automatically enabled (true) and cannot be disabled
 
         model : typing.Optional[TextToSpeechModel]
-            Specifies the model to use for text-to-speech conversion. Default is bulbul:v2.
+            Specifies the model to use for text-to-speech conversion.
+
+            **Available models:**
+            - **bulbul:v2:** Default model with pitch, loudness controls
+            - **bulbul:v3-beta:** Newer model with temperature control, improved quality
 
         output_audio_codec : typing.Optional[TextToSpeechOutputAudioCodec]
             Specifies the audio codec for the output audio file. Different codecs offer various compression and quality characteristics.
 
+        temperature : typing.Optional[float]
+            Controls the randomness of the output. Lower values make the output more focused and deterministic, while higher values make it more random. The suitable range is between 0.01 and 1.0. Default is 0.6.
+
+            **Note:** This parameter is only supported for bulbul:v3-beta. It has no effect on bulbul:v2.
+
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
@@ -110,6 +160,7 @@ class RawTextToSpeechClient:
                 "enable_preprocessing": enable_preprocessing,
                 "model": model,
                 "output_audio_codec": output_audio_codec,
+                "temperature": temperature,
             },
             headers={
                 "content-type": "application/json",
@@ -205,15 +256,38 @@ class AsyncRawTextToSpeechClient:
         enable_preprocessing: typing.Optional[bool] = OMIT,
         model: typing.Optional[TextToSpeechModel] = OMIT,
         output_audio_codec: typing.Optional[TextToSpeechOutputAudioCodec] = OMIT,
+        temperature: typing.Optional[float] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> AsyncHttpResponse[TextToSpeechResponse]:
         """
-        This is the model to convert text into spoken audio.
-        The output is a wave file encoded as a base64 string.
+        Convert text into spoken audio. The output is a wave file encoded as a base64 string.
+
+        **Available Models:**
+        - **bulbul:v2** (default): Supports pitch, loudness, and pace controls
+        - **bulbul:v3-beta**: Newer model with temperature control and improved quality
+
+        **Important Notes for bulbul:v3-beta:**
+        - Pitch and loudness parameters are NOT supported
+        - Pace must be between 0.5 and 2.0
+        - Preprocessing is automatically enabled
+        - Default sample rate is 24000 Hz
+        - Temperature parameter available (0.01-1.0, default 0.6)
 
         Parameters
         ----------
         text : str
+            The text(s) to be converted into speech.
+
+            **Features:**
+            - Supports code-mixed text (English and Indic languages)
+
+            **Model-specific limits:**
+            - **bulbul:v2:** Max 1500 characters
+            - **bulbul:v3-beta:** Max 2500 characters
+
+            **Important Note:**
+            - For numbers larger than 4 digits, use commas (e.g., '10,000' instead of '10000')
+            - This ensures proper pronunciation as a whole number
 
         target_language_code : TextToSpeechLanguage
             The language of the text is BCP-47 format
@@ -221,36 +295,63 @@ class AsyncRawTextToSpeechClient:
         speaker : typing.Optional[TextToSpeechSpeaker]
             The speaker voice to be used for the output audio.
 
-            **Default:** Anushka
+            **Default:** Anushka (for bulbul:v2), Aditya (for bulbul:v3-beta)
 
             **Model Compatibility (Speakers compatible with respective model):**
             - **bulbul:v2:**
               - Female: Anushka, Manisha, Vidya, Arya
               - Male: Abhilash, Karun, Hitesh
+            - **bulbul:v3-beta:**
+              - Aditya, Ritu, Priya, Neha, Rahul, Pooja, Rohan, Simran, Kavya, Amit, Dev, Ishita, Shreya, Ratan, Varun, Manan, Sumit, Roopa, Kabir, Aayan, Shubh, Ashutosh, Advait, Amelia, Sophia
 
             **Note:** Speaker selection must match the chosen model version.
 
         pitch : typing.Optional[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.
 
+            **Note:** This parameter is only supported for bulbul:v2. It is NOT supported for bulbul:v3-beta and will cause a validation error if provided.
+
         pace : typing.Optional[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.
+            Controls the speed of the audio. Lower values result in slower speech, while higher values make it faster. Default is 1.0.
+
+            **Model-specific ranges:**
+            - **bulbul:v2:** 0.3 to 3.0
+            - **bulbul:v3-beta:** 0.5 to 2.0
 
         loudness : typing.Optional[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.
 
+            **Note:** This parameter is only supported for bulbul:v2. It is NOT supported for bulbul:v3-beta and will cause a validation error if provided.
+
         speech_sample_rate : typing.Optional[SpeechSampleRate]
-            Specifies the sample rate of the output audio. Supported values are 8000, 16000, 22050, 24000 Hz. If not provided, the default is 22050 Hz.
+            Specifies the sample rate of the output audio. Supported values are 8000, 16000, 22050, 24000 Hz.
+
+            **Model-specific defaults:**
+            - **bulbul:v2:** Default is 22050 Hz
+            - **bulbul:v3-beta:** Default is 24000 Hz
 
         enable_preprocessing : typing.Optional[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.
+            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.
+
+            **Model-specific behavior:**
+            - **bulbul:v2:** Default is false
+            - **bulbul:v3-beta:** Automatically enabled (true) and cannot be disabled
 
         model : typing.Optional[TextToSpeechModel]
-            Specifies the model to use for text-to-speech conversion. Default is bulbul:v2.
+            Specifies the model to use for text-to-speech conversion.
+
+            **Available models:**
+            - **bulbul:v2:** Default model with pitch, loudness controls
+            - **bulbul:v3-beta:** Newer model with temperature control, improved quality
 
         output_audio_codec : typing.Optional[TextToSpeechOutputAudioCodec]
             Specifies the audio codec for the output audio file. Different codecs offer various compression and quality characteristics.
 
+        temperature : typing.Optional[float]
+            Controls the randomness of the output. Lower values make the output more focused and deterministic, while higher values make it more random. The suitable range is between 0.01 and 1.0. Default is 0.6.
+
+            **Note:** This parameter is only supported for bulbul:v3-beta. It has no effect on bulbul:v2.
+
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
@@ -274,6 +375,7 @@ class AsyncRawTextToSpeechClient:
                 "enable_preprocessing": enable_preprocessing,
                 "model": model,
                 "output_audio_codec": output_audio_codec,
+                "temperature": temperature,
             },
             headers={
                 "content-type": "application/json",

sarvamai/text_to_speech_streaming/__init__.py

@@ -2,6 +2,6 @@
 
 # isort: skip_file
 
-from .types import TextToSpeechStreamingSendCompletionEvent
+from .types import TextToSpeechStreamingModel, TextToSpeechStreamingSendCompletionEvent
 
-__all__ = ["TextToSpeechStreamingSendCompletionEvent"]
+__all__ = ["TextToSpeechStreamingModel", "TextToSpeechStreamingSendCompletionEvent"]

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

@@ -2,6 +2,7 @@
 
 # isort: skip_file
 
+from .text_to_speech_streaming_model import TextToSpeechStreamingModel
 from .text_to_speech_streaming_send_completion_event import TextToSpeechStreamingSendCompletionEvent
 
-__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

@@ -3,7 +3,6 @@
 # isort: skip_file
 
 from .audio_data import AudioData
-from .audio_data_input_audio_codec import AudioDataInputAudioCodec
 from .audio_message import AudioMessage
 from .audio_output import AudioOutput
 from .audio_output_data import AudioOutputData
@@ -26,6 +25,7 @@ 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
@@ -34,6 +34,21 @@ from .connection_sample_rate import ConnectionSampleRate
 from .create_chat_completion_response import CreateChatCompletionResponse
 from .diarized_entry import DiarizedEntry
 from .diarized_transcript import DiarizedTranscript
+from .doc_digitization_create_job_response import DocDigitizationCreateJobResponse
+from .doc_digitization_download_files_response import DocDigitizationDownloadFilesResponse
+from .doc_digitization_error_code import DocDigitizationErrorCode
+from .doc_digitization_error_details import DocDigitizationErrorDetails
+from .doc_digitization_error_message import DocDigitizationErrorMessage
+from .doc_digitization_job_detail import DocDigitizationJobDetail
+from .doc_digitization_job_detail_state import DocDigitizationJobDetailState
+from .doc_digitization_job_parameters import DocDigitizationJobParameters
+from .doc_digitization_job_state import DocDigitizationJobState
+from .doc_digitization_job_status_response import DocDigitizationJobStatusResponse
+from .doc_digitization_output_format import DocDigitizationOutputFormat
+from .doc_digitization_page_error import DocDigitizationPageError
+from .doc_digitization_supported_language import DocDigitizationSupportedLanguage
+from .doc_digitization_upload_files_response import DocDigitizationUploadFilesResponse
+from .doc_digitization_webhook_callback import DocDigitizationWebhookCallback
 from .error_code import ErrorCode
 from .error_data import ErrorData
 from .error_details import ErrorDetails
@@ -54,6 +69,7 @@ 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
@@ -104,7 +120,6 @@ from .transliteration_response import TransliterationResponse
 
 __all__ = [
     "AudioData",
-    "AudioDataInputAudioCodec",
     "AudioMessage",
     "AudioOutput",
     "AudioOutputData",
@@ -125,6 +140,7 @@ __all__ = [
     "ConfigMessage",
     "ConfigureConnection",
     "ConfigureConnectionData",
+    "ConfigureConnectionDataModel",
     "ConfigureConnectionDataOutputAudioBitrate",
     "ConfigureConnectionDataOutputAudioCodec",
     "ConfigureConnectionDataSpeaker",
@@ -133,6 +149,21 @@ __all__ = [
     "CreateChatCompletionResponse",
     "DiarizedEntry",
     "DiarizedTranscript",
+    "DocDigitizationCreateJobResponse",
+    "DocDigitizationDownloadFilesResponse",
+    "DocDigitizationErrorCode",
+    "DocDigitizationErrorDetails",
+    "DocDigitizationErrorMessage",
+    "DocDigitizationJobDetail",
+    "DocDigitizationJobDetailState",
+    "DocDigitizationJobParameters",
+    "DocDigitizationJobState",
+    "DocDigitizationJobStatusResponse",
+    "DocDigitizationOutputFormat",
+    "DocDigitizationPageError",
+    "DocDigitizationSupportedLanguage",
+    "DocDigitizationUploadFilesResponse",
+    "DocDigitizationWebhookCallback",
     "ErrorCode",
     "ErrorData",
     "ErrorDetails",
@@ -153,6 +184,7 @@ __all__ = [
     "JobState",
     "JobStatusV1Response",
     "LanguageIdentificationResponse",
+    "Mode",
     "NumeralsFormat",
     "PingSignal",
     "ReasoningEffort",

sarvamai/types/audio_data.py

@@ -4,7 +4,6 @@ import typing
 
 import pydantic
 from ..core.pydantic_utilities import IS_PYDANTIC_V2, UniversalBaseModel
-from .audio_data_input_audio_codec import AudioDataInputAudioCodec
 
 
 class AudioData(UniversalBaseModel):
@@ -29,11 +28,6 @@ class AudioData(UniversalBaseModel):
     Audio encoding format
     """
 
-    input_audio_codec: typing.Optional[AudioDataInputAudioCodec] = pydantic.Field(default=None)
-    """
-    Audio codec/format of the input file. Our API automatically detects all codec formats, but for PCM files specifically (pcm_s16le, pcm_l16, pcm_raw), you must pass this parameter. PCM files supports sample rate 16000 and 8000.
-    """
-
     if IS_PYDANTIC_V2:
         model_config: typing.ClassVar[pydantic.ConfigDict] = pydantic.ConfigDict(extra="allow", frozen=True)  # type: ignore # Pydantic v2
     else:

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"

sarvamai/types/configure_connection_data.py

@@ -4,6 +4,7 @@ import typing
 
 import pydantic
 from ..core.pydantic_utilities import IS_PYDANTIC_V2, UniversalBaseModel
+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
@@ -11,21 +12,25 @@ from .configure_connection_data_target_language_code import ConfigureConnectionD
 
 
 class ConfigureConnectionData(UniversalBaseModel):
+    model: typing.Optional[ConfigureConnectionDataModel] = pydantic.Field(default=None)
+    """
+    Specifies the model to use for text-to-speech conversion.
+    - **bulbul:v2** (default): Standard TTS model with pitch/loudness support
+    - **bulbul:v3-beta**: Advanced model with temperature control (no pitch/loudness)
+    """
+
     target_language_code: ConfigureConnectionDataTargetLanguageCode = pydantic.Field()
     """
-    The language of the text is BCP-47 format
+    The language of the text in BCP-47 format
     """
 
     speaker: ConfigureConnectionDataSpeaker = 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
+    **Model Compatibility:**
+    - **bulbul:v2:** anushka (default), abhilash, manisha, vidya, arya, karun, hitesh
+    - **bulbul:v3-beta:** aditya (default), ritu, priya, neha, rahul, pooja, rohan, simran, kavya, amit, dev, ishita, shreya, ratan, varun, manan, sumit, roopa, kabir, aayan, shubh, ashutosh, advait, amelia, sophia
     
     **Note:** Speaker selection must match the chosen model version.
     """
@@ -35,13 +40,18 @@ class ConfigureConnectionData(UniversalBaseModel):
     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.
+    
+    **Note:** NOT supported for bulbul:v3-beta. Will be ignored if provided.
     """
 
     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.
+    while higher values make it faster. Default is 1.0.
+    
+    **Model-specific ranges:**
+    - **bulbul:v2:** 0.3 to 3.0
+    - **bulbul:v3-beta:** 0.5 to 2.0
     """
 
     loudness: typing.Optional[float] = pydantic.Field(default=None)
@@ -49,19 +59,38 @@ class ConfigureConnectionData(UniversalBaseModel):
     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.
+    
+    **Note:** NOT supported for bulbul:v3-beta. Will be ignored if provided.
+    """
+
+    temperature: typing.Optional[float] = pydantic.Field(default=None)
+    """
+    Controls the randomness of the output. Lower values make the output more 
+    focused and deterministic, while higher values make it more random. 
+    The suitable range is between 0.01 and 1.0. Default is 0.6.
+    
+    **Note:** Only supported for bulbul:v3-beta. Will be ignored for bulbul:v2.
     """
 
     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.
+    8000, 16000, 22050, 24000 Hz.
+    
+    **Model-specific defaults:**
+    - **bulbul:v2:** 22050 Hz
+    - **bulbul:v3-beta:** 24000 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.
+    of mixed-language text.
+    
+    **Model-specific defaults:**
+    - **bulbul:v2:** false (optional)
+    - **bulbul:v3-beta:** Always enabled (cannot be disabled)
     """
 
     output_audio_codec: typing.Optional[ConfigureConnectionDataOutputAudioCodec] = pydantic.Field(default=None)

sarvamai/types/configure_connection_data_model.py

@@ -0,0 +1,5 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+ConfigureConnectionDataModel = typing.Union[typing.Literal["bulbul:v2", "bulbul:v3-beta"], typing.Any]

sarvamai/types/configure_connection_data_speaker.py

@@ -3,5 +3,39 @@
 import typing
 
 ConfigureConnectionDataSpeaker = typing.Union[
-    typing.Literal["anushka", "abhilash", "manisha", "vidya", "arya", "karun", "hitesh"], typing.Any
+    typing.Literal[
+        "anushka",
+        "abhilash",
+        "manisha",
+        "vidya",
+        "arya",
+        "karun",
+        "hitesh",
+        "aditya",
+        "ritu",
+        "priya",
+        "neha",
+        "rahul",
+        "pooja",
+        "rohan",
+        "simran",
+        "kavya",
+        "amit",
+        "dev",
+        "ishita",
+        "shreya",
+        "ratan",
+        "varun",
+        "manan",
+        "sumit",
+        "roopa",
+        "kabir",
+        "aayan",
+        "shubh",
+        "ashutosh",
+        "advait",
+        "amelia",
+        "sophia",
+    ],
+    typing.Any,
 ]