sarvamai 0.1.23a5__py3-none-any.whl → 0.1.23a7__py3-none-any.whl

sarvamai/requests/__init__.py

@@ -27,6 +27,16 @@ from .configure_connection_data import ConfigureConnectionDataParams
 from .create_chat_completion_response import CreateChatCompletionResponseParams
 from .diarized_entry import DiarizedEntryParams
 from .diarized_transcript import DiarizedTranscriptParams
+from .doc_digitization_create_job_response import DocDigitizationCreateJobResponseParams
+from .doc_digitization_download_files_response import DocDigitizationDownloadFilesResponseParams
+from .doc_digitization_error_details import DocDigitizationErrorDetailsParams
+from .doc_digitization_error_message import DocDigitizationErrorMessageParams
+from .doc_digitization_job_detail import DocDigitizationJobDetailParams
+from .doc_digitization_job_parameters import DocDigitizationJobParametersParams
+from .doc_digitization_job_status_response import DocDigitizationJobStatusResponseParams
+from .doc_digitization_page_error import DocDigitizationPageErrorParams
+from .doc_digitization_upload_files_response import DocDigitizationUploadFilesResponseParams
+from .doc_digitization_webhook_callback import DocDigitizationWebhookCallbackParams
 from .error_data import ErrorDataParams
 from .error_details import ErrorDetailsParams
 from .error_message import ErrorMessageParams
@@ -89,6 +99,16 @@ __all__ = [
     "CreateChatCompletionResponseParams",
     "DiarizedEntryParams",
     "DiarizedTranscriptParams",
+    "DocDigitizationCreateJobResponseParams",
+    "DocDigitizationDownloadFilesResponseParams",
+    "DocDigitizationErrorDetailsParams",
+    "DocDigitizationErrorMessageParams",
+    "DocDigitizationJobDetailParams",
+    "DocDigitizationJobParametersParams",
+    "DocDigitizationJobStatusResponseParams",
+    "DocDigitizationPageErrorParams",
+    "DocDigitizationUploadFilesResponseParams",
+    "DocDigitizationWebhookCallbackParams",
     "ErrorDataParams",
     "ErrorDetailsParams",
     "ErrorMessageParams",

sarvamai/requests/doc_digitization_create_job_response.py

@@ -0,0 +1,25 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing_extensions
+from ..types.doc_digitization_job_state import DocDigitizationJobState
+from ..types.storage_container_type import StorageContainerType
+from .doc_digitization_job_parameters import DocDigitizationJobParametersParams
+
+
+class DocDigitizationCreateJobResponseParams(typing_extensions.TypedDict):
+    job_id: str
+    """
+    Unique job identifier (UUID)
+    """
+
+    storage_container_type: StorageContainerType
+    """
+    Storage Container Type
+    """
+
+    job_parameters: DocDigitizationJobParametersParams
+    """
+    Job configuration parameters
+    """
+
+    job_state: DocDigitizationJobState

sarvamai/requests/doc_digitization_download_files_response.py

@@ -0,0 +1,37 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+import typing_extensions
+from ..types.doc_digitization_job_state import DocDigitizationJobState
+from ..types.storage_container_type import StorageContainerType
+from .file_signed_url_details import FileSignedUrlDetailsParams
+
+
+class DocDigitizationDownloadFilesResponseParams(typing_extensions.TypedDict):
+    """
+    Response for download-files endpoint.
+    """
+
+    job_id: str
+    """
+    Job identifier (UUID)
+    """
+
+    job_state: DocDigitizationJobState
+    """
+    Current job state
+    """
+
+    storage_container_type: StorageContainerType
+    """
+    Storage backend type
+    """
+
+    download_urls: typing.Dict[str, FileSignedUrlDetailsParams]
+    """
+    Map of filename to presigned download URL details
+    """
+
+    error_code: typing_extensions.NotRequired[str]
+    error_message: typing_extensions.NotRequired[str]

sarvamai/requests/doc_digitization_error_details.py

@@ -0,0 +1,21 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing_extensions
+from ..types.doc_digitization_error_code import DocDigitizationErrorCode
+
+
+class DocDigitizationErrorDetailsParams(typing_extensions.TypedDict):
+    message: str
+    """
+    Message describing the error
+    """
+
+    code: DocDigitizationErrorCode
+    """
+    Error code for the specific error that has occurred.
+    """
+
+    request_id: typing_extensions.NotRequired[str]
+    """
+    Unique identifier for the request. Format: date_UUID4
+    """

sarvamai/requests/doc_digitization_error_message.py

@@ -0,0 +1,11 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing_extensions
+from .doc_digitization_error_details import DocDigitizationErrorDetailsParams
+
+
+class DocDigitizationErrorMessageParams(typing_extensions.TypedDict):
+    error: DocDigitizationErrorDetailsParams
+    """
+    Error details
+    """

sarvamai/requests/doc_digitization_job_detail.py

@@ -0,0 +1,64 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+import typing_extensions
+from ..types.doc_digitization_job_detail_state import DocDigitizationJobDetailState
+from .doc_digitization_page_error import DocDigitizationPageErrorParams
+from .task_file_details import TaskFileDetailsParams
+
+
+class DocDigitizationJobDetailParams(typing_extensions.TypedDict):
+    """
+    Processing details for a single input file with page-level metrics.
+    """
+
+    inputs: typing.Sequence[TaskFileDetailsParams]
+    """
+    Input file(s) for this task
+    """
+
+    outputs: typing.Sequence[TaskFileDetailsParams]
+    """
+    Output file(s) produced
+    """
+
+    state: DocDigitizationJobDetailState
+    """
+    Processing state for this file
+    """
+
+    total_pages: typing_extensions.NotRequired[int]
+    """
+    Total pages/images in the input file
+    """
+
+    pages_processed: typing_extensions.NotRequired[int]
+    """
+    Number of pages processed so far
+    """
+
+    pages_succeeded: typing_extensions.NotRequired[int]
+    """
+    Number of pages successfully processed
+    """
+
+    pages_failed: typing_extensions.NotRequired[int]
+    """
+    Number of pages that failed processing
+    """
+
+    error_message: typing_extensions.NotRequired[str]
+    """
+    Error message if processing failed
+    """
+
+    error_code: typing_extensions.NotRequired[str]
+    """
+    Standardized error code if failed
+    """
+
+    page_errors: typing_extensions.NotRequired[typing.Sequence[DocDigitizationPageErrorParams]]
+    """
+    Detailed errors for each failed page
+    """

sarvamai/requests/doc_digitization_job_parameters.py

@@ -0,0 +1,21 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing_extensions
+from ..types.doc_digitization_output_format import DocDigitizationOutputFormat
+from ..types.doc_digitization_supported_language import DocDigitizationSupportedLanguage
+
+
+class DocDigitizationJobParametersParams(typing_extensions.TypedDict):
+    """
+    Job parameters for document digitization.
+    """
+
+    language: typing_extensions.NotRequired[DocDigitizationSupportedLanguage]
+    """
+    ISO language code for the document
+    """
+
+    output_format: typing_extensions.NotRequired[DocDigitizationOutputFormat]
+    """
+    Output format: html or md
+    """

sarvamai/requests/doc_digitization_job_status_response.py

@@ -0,0 +1,65 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import datetime as dt
+import typing
+
+import typing_extensions
+from ..types.doc_digitization_job_state import DocDigitizationJobState
+from ..types.storage_container_type import StorageContainerType
+from .doc_digitization_job_detail import DocDigitizationJobDetailParams
+
+
+class DocDigitizationJobStatusResponseParams(typing_extensions.TypedDict):
+    """
+    Response model for job status endpoint.
+    """
+
+    job_id: str
+    """
+    Job identifier (UUID)
+    """
+
+    job_state: DocDigitizationJobState
+    """
+    Current job state
+    """
+
+    created_at: dt.datetime
+    """
+    Job creation timestamp (ISO 8601)
+    """
+
+    updated_at: dt.datetime
+    """
+    Last update timestamp (ISO 8601)
+    """
+
+    storage_container_type: StorageContainerType
+    """
+    Storage backend type
+    """
+
+    total_files: typing_extensions.NotRequired[int]
+    """
+    Total input files (always 1)
+    """
+
+    successful_files_count: typing_extensions.NotRequired[int]
+    """
+    Files that completed successfully
+    """
+
+    failed_files_count: typing_extensions.NotRequired[int]
+    """
+    Files that failed
+    """
+
+    error_message: typing_extensions.NotRequired[str]
+    """
+    Job-level error message
+    """
+
+    job_details: typing_extensions.NotRequired[typing.Sequence[DocDigitizationJobDetailParams]]
+    """
+    Per-file processing details with page metrics
+    """

sarvamai/requests/doc_digitization_page_error.py

@@ -0,0 +1,24 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing_extensions
+
+
+class DocDigitizationPageErrorParams(typing_extensions.TypedDict):
+    """
+    Error details for a specific page.
+    """
+
+    page_number: int
+    """
+    Page number that failed
+    """
+
+    error_code: str
+    """
+    Standardized error code
+    """
+
+    error_message: str
+    """
+    Human-readable error description
+    """

sarvamai/requests/doc_digitization_upload_files_response.py

@@ -0,0 +1,34 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+import typing_extensions
+from ..types.doc_digitization_job_state import DocDigitizationJobState
+from ..types.storage_container_type import StorageContainerType
+from .file_signed_url_details import FileSignedUrlDetailsParams
+
+
+class DocDigitizationUploadFilesResponseParams(typing_extensions.TypedDict):
+    """
+    Response with presigned upload URLs
+    """
+
+    job_id: str
+    """
+    Job identifier
+    """
+
+    job_state: DocDigitizationJobState
+    """
+    Current job state
+    """
+
+    upload_urls: typing.Dict[str, FileSignedUrlDetailsParams]
+    """
+    Map of filename to presigned upload URL details
+    """
+
+    storage_container_type: StorageContainerType
+    """
+    Storage backend type
+    """

sarvamai/requests/doc_digitization_webhook_callback.py

@@ -0,0 +1,19 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing_extensions
+
+
+class DocDigitizationWebhookCallbackParams(typing_extensions.TypedDict):
+    """
+    Webhook configuration for job completion notification
+    """
+
+    url: str
+    """
+    HTTPS webhook URL to call upon job completion (HTTP not allowed)
+    """
+
+    auth_token: typing_extensions.NotRequired[str]
+    """
+    Authorization token sent as X-SARVAM-JOB-CALLBACK-TOKEN header
+    """

sarvamai/speech_to_text_streaming/__init__.py

@@ -5,6 +5,7 @@
 from .types import (
     SpeechToTextStreamingFlushSignal,
     SpeechToTextStreamingHighVadSensitivity,
+    SpeechToTextStreamingInputAudioCodec,
     SpeechToTextStreamingLanguageCode,
     SpeechToTextStreamingMode,
     SpeechToTextStreamingModel,
@@ -14,6 +15,7 @@ from .types import (
 __all__ = [
     "SpeechToTextStreamingFlushSignal",
     "SpeechToTextStreamingHighVadSensitivity",
+    "SpeechToTextStreamingInputAudioCodec",
     "SpeechToTextStreamingLanguageCode",
     "SpeechToTextStreamingMode",
     "SpeechToTextStreamingModel",

sarvamai/speech_to_text_streaming/client.py

@@ -13,6 +13,7 @@ from .raw_client import AsyncRawSpeechToTextStreamingClient, RawSpeechToTextStre
 from .socket_client import AsyncSpeechToTextStreamingSocketClient, SpeechToTextStreamingSocketClient
 from .types.speech_to_text_streaming_flush_signal import SpeechToTextStreamingFlushSignal
 from .types.speech_to_text_streaming_high_vad_sensitivity import SpeechToTextStreamingHighVadSensitivity
+from .types.speech_to_text_streaming_input_audio_codec import SpeechToTextStreamingInputAudioCodec
 from .types.speech_to_text_streaming_language_code import SpeechToTextStreamingLanguageCode
 from .types.speech_to_text_streaming_mode import SpeechToTextStreamingMode
 from .types.speech_to_text_streaming_model import SpeechToTextStreamingModel
@@ -50,6 +51,7 @@ class SpeechToTextStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextStreamingFlushSignal] = None,
+        input_audio_codec: typing.Optional[SpeechToTextStreamingInputAudioCodec] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[SpeechToTextStreamingSocketClient]:
@@ -117,6 +119,10 @@ class SpeechToTextStreamingClient:
         flush_signal : typing.Optional[SpeechToTextStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription
 
+        input_audio_codec : typing.Optional[SpeechToTextStreamingInputAudioCodec]
+            Audio codec/format of the input stream. Use this when sending raw PCM audio.
+            Supported values: wav, pcm_s16le, pcm_l16, pcm_raw.
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -143,6 +149,8 @@ class SpeechToTextStreamingClient:
             query_params = query_params.add("vad_signals", vad_signals)
         if flush_signal is not None:
             query_params = query_params.add("flush_signal", flush_signal)
+        if input_audio_codec is not None:
+            query_params = query_params.add("input_audio_codec", input_audio_codec)
         ws_url = ws_url + f"?{query_params}"
         headers = self._raw_client._client_wrapper.get_headers()
         if api_subscription_key is not None:
@@ -193,6 +201,7 @@ class AsyncSpeechToTextStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextStreamingFlushSignal] = None,
+        input_audio_codec: typing.Optional[SpeechToTextStreamingInputAudioCodec] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncSpeechToTextStreamingSocketClient]:
@@ -260,6 +269,10 @@ class AsyncSpeechToTextStreamingClient:
         flush_signal : typing.Optional[SpeechToTextStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription
 
+        input_audio_codec : typing.Optional[SpeechToTextStreamingInputAudioCodec]
+            Audio codec/format of the input stream. Use this when sending raw PCM audio.
+            Supported values: wav, pcm_s16le, pcm_l16, pcm_raw.
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -286,6 +299,8 @@ class AsyncSpeechToTextStreamingClient:
             query_params = query_params.add("vad_signals", vad_signals)
         if flush_signal is not None:
             query_params = query_params.add("flush_signal", flush_signal)
+        if input_audio_codec is not None:
+            query_params = query_params.add("input_audio_codec", input_audio_codec)
         ws_url = ws_url + f"?{query_params}"
         headers = self._raw_client._client_wrapper.get_headers()
         if api_subscription_key is not None:

sarvamai/speech_to_text_streaming/raw_client.py

@@ -12,6 +12,7 @@ from ..core.request_options import RequestOptions
 from .socket_client import AsyncSpeechToTextStreamingSocketClient, SpeechToTextStreamingSocketClient
 from .types.speech_to_text_streaming_flush_signal import SpeechToTextStreamingFlushSignal
 from .types.speech_to_text_streaming_high_vad_sensitivity import SpeechToTextStreamingHighVadSensitivity
+from .types.speech_to_text_streaming_input_audio_codec import SpeechToTextStreamingInputAudioCodec
 from .types.speech_to_text_streaming_language_code import SpeechToTextStreamingLanguageCode
 from .types.speech_to_text_streaming_mode import SpeechToTextStreamingMode
 from .types.speech_to_text_streaming_model import SpeechToTextStreamingModel
@@ -38,6 +39,7 @@ class RawSpeechToTextStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextStreamingFlushSignal] = None,
+        input_audio_codec: typing.Optional[SpeechToTextStreamingInputAudioCodec] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[SpeechToTextStreamingSocketClient]:
@@ -105,6 +107,10 @@ class RawSpeechToTextStreamingClient:
         flush_signal : typing.Optional[SpeechToTextStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription
 
+        input_audio_codec : typing.Optional[SpeechToTextStreamingInputAudioCodec]
+            Audio codec/format of the input stream. Use this when sending raw PCM audio.
+            Supported values: wav, pcm_s16le, pcm_l16, pcm_raw.
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -131,6 +137,8 @@ class RawSpeechToTextStreamingClient:
             query_params = query_params.add("vad_signals", vad_signals)
         if flush_signal is not None:
             query_params = query_params.add("flush_signal", flush_signal)
+        if input_audio_codec is not None:
+            query_params = query_params.add("input_audio_codec", input_audio_codec)
         ws_url = ws_url + f"?{query_params}"
         headers = self._client_wrapper.get_headers()
         if api_subscription_key is not None:
@@ -170,6 +178,7 @@ class AsyncRawSpeechToTextStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextStreamingFlushSignal] = None,
+        input_audio_codec: typing.Optional[SpeechToTextStreamingInputAudioCodec] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncSpeechToTextStreamingSocketClient]:
@@ -237,6 +246,10 @@ class AsyncRawSpeechToTextStreamingClient:
         flush_signal : typing.Optional[SpeechToTextStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription
 
+        input_audio_codec : typing.Optional[SpeechToTextStreamingInputAudioCodec]
+            Audio codec/format of the input stream. Use this when sending raw PCM audio.
+            Supported values: wav, pcm_s16le, pcm_l16, pcm_raw.
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -263,6 +276,8 @@ class AsyncRawSpeechToTextStreamingClient:
             query_params = query_params.add("vad_signals", vad_signals)
         if flush_signal is not None:
             query_params = query_params.add("flush_signal", flush_signal)
+        if input_audio_codec is not None:
+            query_params = query_params.add("input_audio_codec", input_audio_codec)
         ws_url = ws_url + f"?{query_params}"
         headers = self._client_wrapper.get_headers()
         if api_subscription_key is not None:

sarvamai/speech_to_text_streaming/types/__init__.py

@@ -4,6 +4,7 @@
 
 from .speech_to_text_streaming_flush_signal import SpeechToTextStreamingFlushSignal
 from .speech_to_text_streaming_high_vad_sensitivity import SpeechToTextStreamingHighVadSensitivity
+from .speech_to_text_streaming_input_audio_codec import SpeechToTextStreamingInputAudioCodec
 from .speech_to_text_streaming_language_code import SpeechToTextStreamingLanguageCode
 from .speech_to_text_streaming_mode import SpeechToTextStreamingMode
 from .speech_to_text_streaming_model import SpeechToTextStreamingModel
@@ -12,6 +13,7 @@ from .speech_to_text_streaming_vad_signals import SpeechToTextStreamingVadSignal
 __all__ = [
     "SpeechToTextStreamingFlushSignal",
     "SpeechToTextStreamingHighVadSensitivity",
+    "SpeechToTextStreamingInputAudioCodec",
     "SpeechToTextStreamingLanguageCode",
     "SpeechToTextStreamingMode",
     "SpeechToTextStreamingModel",

sarvamai/speech_to_text_streaming/types/speech_to_text_streaming_input_audio_codec.py

@@ -0,0 +1,7 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+SpeechToTextStreamingInputAudioCodec = typing.Union[
+    typing.Literal["wav", "pcm_s16le", "pcm_l16", "pcm_raw"], typing.Any
+]

sarvamai/speech_to_text_translate_streaming/__init__.py

@@ -5,11 +5,13 @@
 from .types import (
     SpeechToTextTranslateStreamingFlushSignal,
     SpeechToTextTranslateStreamingHighVadSensitivity,
+    SpeechToTextTranslateStreamingInputAudioCodec,
     SpeechToTextTranslateStreamingVadSignals,
 )
 
 __all__ = [
     "SpeechToTextTranslateStreamingFlushSignal",
     "SpeechToTextTranslateStreamingHighVadSensitivity",
+    "SpeechToTextTranslateStreamingInputAudioCodec",
     "SpeechToTextTranslateStreamingVadSignals",
 ]

sarvamai/speech_to_text_translate_streaming/client.py

@@ -15,6 +15,7 @@ from .types.speech_to_text_translate_streaming_flush_signal import SpeechToTextT
 from .types.speech_to_text_translate_streaming_high_vad_sensitivity import (
     SpeechToTextTranslateStreamingHighVadSensitivity,
 )
+from .types.speech_to_text_translate_streaming_input_audio_codec import SpeechToTextTranslateStreamingInputAudioCodec
 from .types.speech_to_text_translate_streaming_vad_signals import SpeechToTextTranslateStreamingVadSignals
 
 try:
@@ -47,6 +48,7 @@ class SpeechToTextTranslateStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextTranslateStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextTranslateStreamingFlushSignal] = None,
+        input_audio_codec: typing.Optional[SpeechToTextTranslateStreamingInputAudioCodec] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[SpeechToTextTranslateStreamingSocketClient]:
@@ -77,6 +79,10 @@ class SpeechToTextTranslateStreamingClient:
         flush_signal : typing.Optional[SpeechToTextTranslateStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription and translation
 
+        input_audio_codec : typing.Optional[SpeechToTextTranslateStreamingInputAudioCodec]
+            Audio codec/format of the input stream. Use this when sending raw PCM audio.
+            Supported values: wav, pcm_s16le, pcm_l16, pcm_raw.
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -99,6 +105,8 @@ class SpeechToTextTranslateStreamingClient:
             query_params = query_params.add("vad_signals", vad_signals)
         if flush_signal is not None:
             query_params = query_params.add("flush_signal", flush_signal)
+        if input_audio_codec is not None:
+            query_params = query_params.add("input_audio_codec", input_audio_codec)
         ws_url = ws_url + f"?{query_params}"
         headers = self._raw_client._client_wrapper.get_headers()
         if api_subscription_key is not None:
@@ -147,6 +155,7 @@ class AsyncSpeechToTextTranslateStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextTranslateStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextTranslateStreamingFlushSignal] = None,
+        input_audio_codec: typing.Optional[SpeechToTextTranslateStreamingInputAudioCodec] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncSpeechToTextTranslateStreamingSocketClient]:
@@ -177,6 +186,10 @@ class AsyncSpeechToTextTranslateStreamingClient:
         flush_signal : typing.Optional[SpeechToTextTranslateStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription and translation
 
+        input_audio_codec : typing.Optional[SpeechToTextTranslateStreamingInputAudioCodec]
+            Audio codec/format of the input stream. Use this when sending raw PCM audio.
+            Supported values: wav, pcm_s16le, pcm_l16, pcm_raw.
+
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -199,6 +212,8 @@ class AsyncSpeechToTextTranslateStreamingClient:
             query_params = query_params.add("vad_signals", vad_signals)
         if flush_signal is not None:
             query_params = query_params.add("flush_signal", flush_signal)
+        if input_audio_codec is not None:
+            query_params = query_params.add("input_audio_codec", input_audio_codec)
         ws_url = ws_url + f"?{query_params}"
         headers = self._raw_client._client_wrapper.get_headers()
         if api_subscription_key is not None: