sarvamai 0.1.22a3__py3-none-any.whl → 0.1.22a4__py3-none-any.whl

sarvamai/__init__.py

@@ -56,7 +56,6 @@ from .types import (
     NumeralsFormat,
     PingSignal,
     ReasoningEffort,
-    ResponseSpeechState,
     ResponseType,
     Role,
     SarvamModelIds,
@@ -190,14 +189,12 @@ from .speech_to_text_streaming import (
     SpeechToTextStreamingHighVadSensitivity,
     SpeechToTextStreamingInputAudioCodec,
     SpeechToTextStreamingLanguageCode,
-    SpeechToTextStreamingStreamOngoingSpeechResults,
     SpeechToTextStreamingVadSignals,
 )
 from .speech_to_text_translate_streaming import (
     SpeechToTextTranslateStreamingFlushSignal,
     SpeechToTextTranslateStreamingHighVadSensitivity,
     SpeechToTextTranslateStreamingInputAudioCodec,
-    SpeechToTextTranslateStreamingStreamOngoingSpeechResults,
     SpeechToTextTranslateStreamingVadSignals,
 )
 from .text_to_speech_streaming import TextToSpeechStreamingSendCompletionEvent
@@ -300,7 +297,6 @@ __all__ = [
     "PingSignal",
     "PingSignalParams",
     "ReasoningEffort",
-    "ResponseSpeechState",
     "ResponseType",
     "Role",
     "SarvamAI",
@@ -326,7 +322,6 @@ __all__ = [
     "SpeechToTextStreamingLanguageCode",
     "SpeechToTextStreamingResponse",
     "SpeechToTextStreamingResponseParams",
-    "SpeechToTextStreamingStreamOngoingSpeechResults",
     "SpeechToTextStreamingVadSignals",
     "SpeechToTextTranscriptionData",
     "SpeechToTextTranscriptionDataParams",
@@ -343,7 +338,6 @@ __all__ = [
     "SpeechToTextTranslateStreamingInputAudioCodec",
     "SpeechToTextTranslateStreamingResponse",
     "SpeechToTextTranslateStreamingResponseParams",
-    "SpeechToTextTranslateStreamingStreamOngoingSpeechResults",
     "SpeechToTextTranslateStreamingVadSignals",
     "SpeechToTextTranslateTranscriptionData",
     "SpeechToTextTranslateTranscriptionDataParams",

sarvamai/core/client_wrapper.py

@@ -23,10 +23,10 @@ class BaseClientWrapper:
 
     def get_headers(self) -> typing.Dict[str, str]:
         headers: typing.Dict[str, str] = {
-            "User-Agent": "sarvamai/0.1.22a3",
+            "User-Agent": "sarvamai/0.1.22a4",
             "X-Fern-Language": "Python",
             "X-Fern-SDK-Name": "sarvamai",
-            "X-Fern-SDK-Version": "0.1.22a3",
+            "X-Fern-SDK-Version": "0.1.22a4",
             **(self.get_custom_headers() or {}),
         }
         headers["api-subscription-key"] = self.api_subscription_key

sarvamai/requests/speech_to_text_transcription_data.py

@@ -3,7 +3,6 @@
 import typing
 
 import typing_extensions
-from ..types.response_speech_state import ResponseSpeechState
 from .transcription_metrics import TranscriptionMetricsParams
 
 
@@ -33,9 +32,4 @@ class SpeechToTextTranscriptionDataParams(typing_extensions.TypedDict):
     BCP-47 code of detected language
     """
 
-    response_speech_state: typing_extensions.NotRequired[ResponseSpeechState]
-    """
-    Current state of speech detection and processing
-    """
-
     metrics: TranscriptionMetricsParams

sarvamai/requests/speech_to_text_translate_transcription_data.py

@@ -1,7 +1,6 @@
 # This file was auto-generated by Fern from our API Definition.
 
 import typing_extensions
-from ..types.response_speech_state import ResponseSpeechState
 from .transcription_metrics import TranscriptionMetricsParams
 
 
@@ -21,9 +20,4 @@ class SpeechToTextTranslateTranscriptionDataParams(typing_extensions.TypedDict):
     BCP-47 code of detected source language (null when language detection is in progress)
     """
 
-    response_speech_state: typing_extensions.NotRequired[ResponseSpeechState]
-    """
-    Current state of speech detection and processing
-    """
-
     metrics: TranscriptionMetricsParams

sarvamai/speech_to_text_job/job.py

@@ -146,9 +146,58 @@ class AsyncSpeechToTextJob:
                 "output_file": detail.outputs[0].file_name,
             }
             for detail in (job_status.job_details or [])
-            if detail.inputs and detail.outputs
+            if detail.inputs and detail.outputs and detail.state == "Success"
         ]
 
+    async def get_file_results(
+        self,
+    ) -> typing.Dict[str, typing.List[typing.Dict[str, typing.Any]]]:
+        """
+        Get detailed results for each file in the batch job.
+
+        Returns
+        -------
+        Dict[str, List[Dict[str, Any]]]
+            Dictionary with 'successful' and 'failed' keys, each containing a list of file details.
+            Each file detail includes:
+            - 'file_name': Name of the input file
+            - 'status': Status of processing ('Success' or 'Failed')
+            - 'error_message': Error message if failed (None if successful)
+            - 'output_file': Name of output file if successful (None if failed)
+        """
+        job_status = await self.get_status()
+        results: typing.Dict[str, typing.List[typing.Dict[str, typing.Any]]] = {
+            "successful": [],
+            "failed": [],
+        }
+
+        for detail in job_status.job_details or []:
+            # Check for empty lists explicitly
+            if not detail.inputs or len(detail.inputs) == 0:
+                continue
+
+            try:
+                file_info = {
+                    "file_name": detail.inputs[0].file_name,
+                    "status": detail.state,
+                    "error_message": detail.error_message,
+                    "output_file": (
+                        detail.outputs[0].file_name
+                        if detail.outputs and len(detail.outputs) > 0
+                        else None
+                    ),
+                }
+
+                if detail.state == "Success":
+                    results["successful"].append(file_info)
+                else:
+                    results["failed"].append(file_info)
+            except (IndexError, AttributeError):
+                # Skip malformed job details
+                continue
+
+        return results
+
     async def download_outputs(self, output_dir: str) -> bool:
         """
         Download output files to the specified directory.
@@ -387,9 +436,58 @@ class SpeechToTextJob:
                 "output_file": detail.outputs[0].file_name,
             }
             for detail in (job_status.job_details or [])
-            if detail.inputs and detail.outputs
+            if detail.inputs and detail.outputs and detail.state == "Success"
         ]
 
+    def get_file_results(
+        self,
+    ) -> typing.Dict[str, typing.List[typing.Dict[str, typing.Any]]]:
+        """
+        Get detailed results for each file in the batch job.
+
+        Returns
+        -------
+        Dict[str, List[Dict[str, Any]]]
+            Dictionary with 'successful' and 'failed' keys, each containing a list of file details.
+            Each file detail includes:
+            - 'file_name': Name of the input file
+            - 'status': Status of processing ('Success' or 'Failed')
+            - 'error_message': Error message if failed (None if successful)
+            - 'output_file': Name of output file if successful (None if failed)
+        """
+        job_status = self.get_status()
+        results: typing.Dict[str, typing.List[typing.Dict[str, typing.Any]]] = {
+            "successful": [],
+            "failed": [],
+        }
+
+        for detail in job_status.job_details or []:
+            # Check for empty lists explicitly
+            if not detail.inputs or len(detail.inputs) == 0:
+                continue
+
+            try:
+                file_info = {
+                    "file_name": detail.inputs[0].file_name,
+                    "status": detail.state,
+                    "error_message": detail.error_message,
+                    "output_file": (
+                        detail.outputs[0].file_name
+                        if detail.outputs and len(detail.outputs) > 0
+                        else None
+                    ),
+                }
+
+                if detail.state == "Success":
+                    results["successful"].append(file_info)
+                else:
+                    results["failed"].append(file_info)
+            except (IndexError, AttributeError):
+                # Skip malformed job details
+                continue
+
+        return results
+
     def download_outputs(self, output_dir: str) -> bool:
         """
         Download output files to the specified directory.

sarvamai/speech_to_text_job/raw_client.py

@@ -39,7 +39,7 @@ class RawSpeechToTextJobClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> HttpResponse[BulkJobInitResponseV1]:
         """
-        Get a job uuid, and storage folder details for speech to text bulk job v1
+        Create a new speech to text bulk job and receive a job UUID and storage folder details for processing multiple audio files
 
         Parameters
         ----------
@@ -160,7 +160,9 @@ class RawSpeechToTextJobClient:
         self, job_id: str, *, request_options: typing.Optional[RequestOptions] = None
     ) -> HttpResponse[JobStatusV1Response]:
         """
-        Get the status of a speech to text bulk job V1
+        Retrieve the current status and details of a speech to text bulk job, including progress and file-level information.
+
+        **Rate Limiting Best Practice:** To prevent rate limit errors and ensure optimal server performance, we recommend implementing a minimum 5-millisecond delay between consecutive status polling requests. This helps maintain system stability while still providing timely status updates.
 
         Parameters
         ----------
@@ -270,7 +272,7 @@ class RawSpeechToTextJobClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> HttpResponse[JobStatusV1Response]:
         """
-        Start a speech to text bulk job V1
+        Start processing a speech to text bulk job after all audio files have been uploaded
 
         Parameters
         ----------
@@ -381,7 +383,7 @@ class RawSpeechToTextJobClient:
         self, *, job_id: str, files: typing.Sequence[str], request_options: typing.Optional[RequestOptions] = None
     ) -> HttpResponse[FilesUploadResponse]:
         """
-        Start a speech to text bulk job V1
+        Generate presigned upload URLs for audio files that will be processed in a speech to text bulk job
 
         Parameters
         ----------
@@ -496,7 +498,7 @@ class RawSpeechToTextJobClient:
         self, *, job_id: str, files: typing.Sequence[str], request_options: typing.Optional[RequestOptions] = None
     ) -> HttpResponse[FilesDownloadResponse]:
         """
-        Start a speech to text bulk job V1
+        Generate presigned download URLs for the transcription output files of a completed speech to text bulk job
 
         Parameters
         ----------
@@ -620,7 +622,7 @@ class AsyncRawSpeechToTextJobClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> AsyncHttpResponse[BulkJobInitResponseV1]:
         """
-        Get a job uuid, and storage folder details for speech to text bulk job v1
+        Create a new speech to text bulk job and receive a job UUID and storage folder details for processing multiple audio files
 
         Parameters
         ----------
@@ -741,7 +743,9 @@ class AsyncRawSpeechToTextJobClient:
         self, job_id: str, *, request_options: typing.Optional[RequestOptions] = None
     ) -> AsyncHttpResponse[JobStatusV1Response]:
         """
-        Get the status of a speech to text bulk job V1
+        Retrieve the current status and details of a speech to text bulk job, including progress and file-level information.
+
+        **Rate Limiting Best Practice:** To prevent rate limit errors and ensure optimal server performance, we recommend implementing a minimum 5-millisecond delay between consecutive status polling requests. This helps maintain system stability while still providing timely status updates.
 
         Parameters
         ----------
@@ -851,7 +855,7 @@ class AsyncRawSpeechToTextJobClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> AsyncHttpResponse[JobStatusV1Response]:
         """
-        Start a speech to text bulk job V1
+        Start processing a speech to text bulk job after all audio files have been uploaded
 
         Parameters
         ----------
@@ -962,7 +966,7 @@ class AsyncRawSpeechToTextJobClient:
         self, *, job_id: str, files: typing.Sequence[str], request_options: typing.Optional[RequestOptions] = None
     ) -> AsyncHttpResponse[FilesUploadResponse]:
         """
-        Start a speech to text bulk job V1
+        Generate presigned upload URLs for audio files that will be processed in a speech to text bulk job
 
         Parameters
         ----------
@@ -1077,7 +1081,7 @@ class AsyncRawSpeechToTextJobClient:
         self, *, job_id: str, files: typing.Sequence[str], request_options: typing.Optional[RequestOptions] = None
     ) -> AsyncHttpResponse[FilesDownloadResponse]:
         """
-        Start a speech to text bulk job V1
+        Generate presigned download URLs for the transcription output files of a completed speech to text bulk job
 
         Parameters
         ----------

sarvamai/speech_to_text_streaming/__init__.py

@@ -7,7 +7,6 @@ from .types import (
     SpeechToTextStreamingHighVadSensitivity,
     SpeechToTextStreamingInputAudioCodec,
     SpeechToTextStreamingLanguageCode,
-    SpeechToTextStreamingStreamOngoingSpeechResults,
     SpeechToTextStreamingVadSignals,
 )
 
@@ -16,6 +15,5 @@ __all__ = [
     "SpeechToTextStreamingHighVadSensitivity",
     "SpeechToTextStreamingInputAudioCodec",
     "SpeechToTextStreamingLanguageCode",
-    "SpeechToTextStreamingStreamOngoingSpeechResults",
     "SpeechToTextStreamingVadSignals",
 ]

sarvamai/speech_to_text_streaming/client.py

@@ -15,9 +15,6 @@ from .types.speech_to_text_streaming_flush_signal import SpeechToTextStreamingFl
 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_stream_ongoing_speech_results import (
-    SpeechToTextStreamingStreamOngoingSpeechResults,
-)
 from .types.speech_to_text_streaming_vad_signals import SpeechToTextStreamingVadSignals
 
 try:
@@ -52,8 +49,6 @@ class SpeechToTextStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextStreamingFlushSignal] = None,
-        stream_ongoing_speech_results: typing.Optional[SpeechToTextStreamingStreamOngoingSpeechResults] = None,
-        streaming_ongoing_requests_frame_size: typing.Optional[str] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[SpeechToTextStreamingSocketClient]:
@@ -87,12 +82,6 @@ class SpeechToTextStreamingClient:
         flush_signal : typing.Optional[SpeechToTextStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription
 
-        stream_ongoing_speech_results : typing.Optional[SpeechToTextStreamingStreamOngoingSpeechResults]
-            Enable streaming of ongoing speech results during active speech
-
-        streaming_ongoing_requests_frame_size : typing.Optional[str]
-            Frame size for streaming ongoing speech results (1-100)
-
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -119,12 +108,6 @@ 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 stream_ongoing_speech_results is not None:
-            query_params = query_params.add("stream_ongoing_speech_results", stream_ongoing_speech_results)
-        if streaming_ongoing_requests_frame_size is not None:
-            query_params = query_params.add(
-                "streaming_ongoing_requests_frame_size", streaming_ongoing_requests_frame_size
-            )
         ws_url = ws_url + f"?{query_params}"
         headers = self._raw_client._client_wrapper.get_headers()
         if api_subscription_key is not None:
@@ -175,8 +158,6 @@ class AsyncSpeechToTextStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextStreamingFlushSignal] = None,
-        stream_ongoing_speech_results: typing.Optional[SpeechToTextStreamingStreamOngoingSpeechResults] = None,
-        streaming_ongoing_requests_frame_size: typing.Optional[str] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncSpeechToTextStreamingSocketClient]:
@@ -210,12 +191,6 @@ class AsyncSpeechToTextStreamingClient:
         flush_signal : typing.Optional[SpeechToTextStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription
 
-        stream_ongoing_speech_results : typing.Optional[SpeechToTextStreamingStreamOngoingSpeechResults]
-            Enable streaming of ongoing speech results during active speech
-
-        streaming_ongoing_requests_frame_size : typing.Optional[str]
-            Frame size for streaming ongoing speech results (1-100)
-
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -242,12 +217,6 @@ 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 stream_ongoing_speech_results is not None:
-            query_params = query_params.add("stream_ongoing_speech_results", stream_ongoing_speech_results)
-        if streaming_ongoing_requests_frame_size is not None:
-            query_params = query_params.add(
-                "streaming_ongoing_requests_frame_size", streaming_ongoing_requests_frame_size
-            )
         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

@@ -14,9 +14,6 @@ from .types.speech_to_text_streaming_flush_signal import SpeechToTextStreamingFl
 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_stream_ongoing_speech_results import (
-    SpeechToTextStreamingStreamOngoingSpeechResults,
-)
 from .types.speech_to_text_streaming_vad_signals import SpeechToTextStreamingVadSignals
 
 try:
@@ -40,8 +37,6 @@ class RawSpeechToTextStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextStreamingFlushSignal] = None,
-        stream_ongoing_speech_results: typing.Optional[SpeechToTextStreamingStreamOngoingSpeechResults] = None,
-        streaming_ongoing_requests_frame_size: typing.Optional[str] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[SpeechToTextStreamingSocketClient]:
@@ -75,12 +70,6 @@ class RawSpeechToTextStreamingClient:
         flush_signal : typing.Optional[SpeechToTextStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription
 
-        stream_ongoing_speech_results : typing.Optional[SpeechToTextStreamingStreamOngoingSpeechResults]
-            Enable streaming of ongoing speech results during active speech
-
-        streaming_ongoing_requests_frame_size : typing.Optional[str]
-            Frame size for streaming ongoing speech results (1-100)
-
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -107,12 +96,6 @@ 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 stream_ongoing_speech_results is not None:
-            query_params = query_params.add("stream_ongoing_speech_results", stream_ongoing_speech_results)
-        if streaming_ongoing_requests_frame_size is not None:
-            query_params = query_params.add(
-                "streaming_ongoing_requests_frame_size", streaming_ongoing_requests_frame_size
-            )
         ws_url = ws_url + f"?{query_params}"
         headers = self._client_wrapper.get_headers()
         if api_subscription_key is not None:
@@ -152,8 +135,6 @@ class AsyncRawSpeechToTextStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextStreamingFlushSignal] = None,
-        stream_ongoing_speech_results: typing.Optional[SpeechToTextStreamingStreamOngoingSpeechResults] = None,
-        streaming_ongoing_requests_frame_size: typing.Optional[str] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncSpeechToTextStreamingSocketClient]:
@@ -187,12 +168,6 @@ class AsyncRawSpeechToTextStreamingClient:
         flush_signal : typing.Optional[SpeechToTextStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription
 
-        stream_ongoing_speech_results : typing.Optional[SpeechToTextStreamingStreamOngoingSpeechResults]
-            Enable streaming of ongoing speech results during active speech
-
-        streaming_ongoing_requests_frame_size : typing.Optional[str]
-            Frame size for streaming ongoing speech results (1-100)
-
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -219,12 +194,6 @@ 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 stream_ongoing_speech_results is not None:
-            query_params = query_params.add("stream_ongoing_speech_results", stream_ongoing_speech_results)
-        if streaming_ongoing_requests_frame_size is not None:
-            query_params = query_params.add(
-                "streaming_ongoing_requests_frame_size", streaming_ongoing_requests_frame_size
-            )
         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

@@ -6,7 +6,6 @@ from .speech_to_text_streaming_flush_signal import SpeechToTextStreamingFlushSig
 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_stream_ongoing_speech_results import SpeechToTextStreamingStreamOngoingSpeechResults
 from .speech_to_text_streaming_vad_signals import SpeechToTextStreamingVadSignals
 
 __all__ = [
@@ -14,6 +13,5 @@ __all__ = [
     "SpeechToTextStreamingHighVadSensitivity",
     "SpeechToTextStreamingInputAudioCodec",
     "SpeechToTextStreamingLanguageCode",
-    "SpeechToTextStreamingStreamOngoingSpeechResults",
     "SpeechToTextStreamingVadSignals",
 ]

sarvamai/speech_to_text_translate_job/job.py

@@ -150,9 +150,58 @@ class AsyncSpeechToTextTranslateJob:
                 "output_file": detail.outputs[0].file_name,
             }
             for detail in (job_status.job_details or [])
-            if detail.inputs and detail.outputs
+            if detail.inputs and detail.outputs and detail.state == "Success"
         ]
 
+    async def get_file_results(
+        self,
+    ) -> typing.Dict[str, typing.List[typing.Dict[str, typing.Any]]]:
+        """
+        Get detailed results for each file in the batch job.
+
+        Returns
+        -------
+        Dict[str, List[Dict[str, Any]]]
+            Dictionary with 'successful' and 'failed' keys, each containing a list of file details.
+            Each file detail includes:
+            - 'file_name': Name of the input file
+            - 'status': Status of processing ('Success' or 'Failed')
+            - 'error_message': Error message if failed (None if successful)
+            - 'output_file': Name of output file if successful (None if failed)
+        """
+        job_status = await self.get_status()
+        results: typing.Dict[str, typing.List[typing.Dict[str, typing.Any]]] = {
+            "successful": [],
+            "failed": [],
+        }
+
+        for detail in job_status.job_details or []:
+            # Check for empty lists explicitly
+            if not detail.inputs or len(detail.inputs) == 0:
+                continue
+
+            try:
+                file_info = {
+                    "file_name": detail.inputs[0].file_name,
+                    "status": detail.state,
+                    "error_message": detail.error_message,
+                    "output_file": (
+                        detail.outputs[0].file_name
+                        if detail.outputs and len(detail.outputs) > 0
+                        else None
+                    ),
+                }
+
+                if detail.state == "Success":
+                    results["successful"].append(file_info)
+                else:
+                    results["failed"].append(file_info)
+            except (IndexError, AttributeError):
+                # Skip malformed job details
+                continue
+
+        return results
+
     async def download_outputs(self, output_dir: str) -> bool:
         """
         Download output files to the specified directory.
@@ -395,9 +444,58 @@ class SpeechToTextTranslateJob:
                 "output_file": detail.outputs[0].file_name,
             }
             for detail in (job_status.job_details or [])
-            if detail.inputs and detail.outputs
+            if detail.inputs and detail.outputs and detail.state == "Success"
         ]
 
+    def get_file_results(
+        self,
+    ) -> typing.Dict[str, typing.List[typing.Dict[str, typing.Any]]]:
+        """
+        Get detailed results for each file in the batch job.
+
+        Returns
+        -------
+        Dict[str, List[Dict[str, Any]]]
+            Dictionary with 'successful' and 'failed' keys, each containing a list of file details.
+            Each file detail includes:
+            - 'file_name': Name of the input file
+            - 'status': Status of processing ('Success' or 'Failed')
+            - 'error_message': Error message if failed (None if successful)
+            - 'output_file': Name of output file if successful (None if failed)
+        """
+        job_status = self.get_status()
+        results: typing.Dict[str, typing.List[typing.Dict[str, typing.Any]]] = {
+            "successful": [],
+            "failed": [],
+        }
+
+        for detail in job_status.job_details or []:
+            # Check for empty lists explicitly
+            if not detail.inputs or len(detail.inputs) == 0:
+                continue
+
+            try:
+                file_info = {
+                    "file_name": detail.inputs[0].file_name,
+                    "status": detail.state,
+                    "error_message": detail.error_message,
+                    "output_file": (
+                        detail.outputs[0].file_name
+                        if detail.outputs and len(detail.outputs) > 0
+                        else None
+                    ),
+                }
+
+                if detail.state == "Success":
+                    results["successful"].append(file_info)
+                else:
+                    results["failed"].append(file_info)
+            except (IndexError, AttributeError):
+                # Skip malformed job details
+                continue
+
+        return results
+
     def download_outputs(self, output_dir: str) -> bool:
         """
         Download output files to the specified directory.

sarvamai/speech_to_text_translate_job/raw_client.py

@@ -40,7 +40,7 @@ class RawSpeechToTextTranslateJobClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> HttpResponse[BulkJobInitResponseV1]:
         """
-        Get a job uuid, and storage folder details for speech to text tranlsate bulk job v1
+        Create a new speech to text translate bulk job and receive a job UUID and storage folder details for processing multiple audio files with translation
 
         Parameters
         ----------
@@ -166,7 +166,9 @@ class RawSpeechToTextTranslateJobClient:
         self, job_id: str, *, request_options: typing.Optional[RequestOptions] = None
     ) -> HttpResponse[JobStatusV1Response]:
         """
-        Get the status of a speech to text translate bulk job V1
+        Retrieve the current status and details of a speech to text translate bulk job, including progress and file-level information.
+
+        **Rate Limiting Best Practice:** To prevent rate limit errors and ensure optimal server performance, we recommend implementing a minimum 5-millisecond delay between consecutive status polling requests. This helps maintain system stability while still providing timely status updates.
 
         Parameters
         ----------
@@ -276,7 +278,7 @@ class RawSpeechToTextTranslateJobClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> HttpResponse[JobStatusV1Response]:
         """
-        Start a speech to text translate bulk job V1
+        Start processing a speech to text translate bulk job after all audio files have been uploaded
 
         Parameters
         ----------
@@ -392,7 +394,7 @@ class RawSpeechToTextTranslateJobClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> HttpResponse[FilesUploadResponse]:
         """
-        Start a speech to text bulk job V1
+        Generate presigned upload URLs for audio files that will be processed in a speech to text translate bulk job
 
         Parameters
         ----------
@@ -517,7 +519,7 @@ class RawSpeechToTextTranslateJobClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> HttpResponse[FilesDownloadResponse]:
         """
-        Start a speech to text bulk job V1
+        Generate presigned download URLs for the translated transcription output files of a completed speech to text translate bulk job
 
         Parameters
         ----------
@@ -647,7 +649,7 @@ class AsyncRawSpeechToTextTranslateJobClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> AsyncHttpResponse[BulkJobInitResponseV1]:
         """
-        Get a job uuid, and storage folder details for speech to text tranlsate bulk job v1
+        Create a new speech to text translate bulk job and receive a job UUID and storage folder details for processing multiple audio files with translation
 
         Parameters
         ----------
@@ -773,7 +775,9 @@ class AsyncRawSpeechToTextTranslateJobClient:
         self, job_id: str, *, request_options: typing.Optional[RequestOptions] = None
     ) -> AsyncHttpResponse[JobStatusV1Response]:
         """
-        Get the status of a speech to text translate bulk job V1
+        Retrieve the current status and details of a speech to text translate bulk job, including progress and file-level information.
+
+        **Rate Limiting Best Practice:** To prevent rate limit errors and ensure optimal server performance, we recommend implementing a minimum 5-millisecond delay between consecutive status polling requests. This helps maintain system stability while still providing timely status updates.
 
         Parameters
         ----------
@@ -883,7 +887,7 @@ class AsyncRawSpeechToTextTranslateJobClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> AsyncHttpResponse[JobStatusV1Response]:
         """
-        Start a speech to text translate bulk job V1
+        Start processing a speech to text translate bulk job after all audio files have been uploaded
 
         Parameters
         ----------
@@ -999,7 +1003,7 @@ class AsyncRawSpeechToTextTranslateJobClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> AsyncHttpResponse[FilesUploadResponse]:
         """
-        Start a speech to text bulk job V1
+        Generate presigned upload URLs for audio files that will be processed in a speech to text translate bulk job
 
         Parameters
         ----------
@@ -1124,7 +1128,7 @@ class AsyncRawSpeechToTextTranslateJobClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> AsyncHttpResponse[FilesDownloadResponse]:
         """
-        Start a speech to text bulk job V1
+        Generate presigned download URLs for the translated transcription output files of a completed speech to text translate bulk job
 
         Parameters
         ----------

sarvamai/speech_to_text_translate_streaming/__init__.py

@@ -6,7 +6,6 @@ from .types import (
     SpeechToTextTranslateStreamingFlushSignal,
     SpeechToTextTranslateStreamingHighVadSensitivity,
     SpeechToTextTranslateStreamingInputAudioCodec,
-    SpeechToTextTranslateStreamingStreamOngoingSpeechResults,
     SpeechToTextTranslateStreamingVadSignals,
 )
 
@@ -14,6 +13,5 @@ __all__ = [
     "SpeechToTextTranslateStreamingFlushSignal",
     "SpeechToTextTranslateStreamingHighVadSensitivity",
     "SpeechToTextTranslateStreamingInputAudioCodec",
-    "SpeechToTextTranslateStreamingStreamOngoingSpeechResults",
     "SpeechToTextTranslateStreamingVadSignals",
 ]

sarvamai/speech_to_text_translate_streaming/client.py

@@ -16,9 +16,6 @@ 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_stream_ongoing_speech_results import (
-    SpeechToTextTranslateStreamingStreamOngoingSpeechResults,
-)
 from .types.speech_to_text_translate_streaming_vad_signals import SpeechToTextTranslateStreamingVadSignals
 
 try:
@@ -52,8 +49,6 @@ class SpeechToTextTranslateStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextTranslateStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextTranslateStreamingFlushSignal] = None,
-        stream_ongoing_speech_results: typing.Optional[SpeechToTextTranslateStreamingStreamOngoingSpeechResults] = None,
-        streaming_ongoing_requests_frame_size: typing.Optional[str] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[SpeechToTextTranslateStreamingSocketClient]:
@@ -84,12 +79,6 @@ class SpeechToTextTranslateStreamingClient:
         flush_signal : typing.Optional[SpeechToTextTranslateStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription and translation
 
-        stream_ongoing_speech_results : typing.Optional[SpeechToTextTranslateStreamingStreamOngoingSpeechResults]
-            Enable streaming of ongoing speech results during active speech
-
-        streaming_ongoing_requests_frame_size : typing.Optional[str]
-            Frame size for streaming ongoing speech results (1-100)
-
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -114,12 +103,6 @@ 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 stream_ongoing_speech_results is not None:
-            query_params = query_params.add("stream_ongoing_speech_results", stream_ongoing_speech_results)
-        if streaming_ongoing_requests_frame_size is not None:
-            query_params = query_params.add(
-                "streaming_ongoing_requests_frame_size", streaming_ongoing_requests_frame_size
-            )
         ws_url = ws_url + f"?{query_params}"
         headers = self._raw_client._client_wrapper.get_headers()
         if api_subscription_key is not None:
@@ -169,8 +152,6 @@ class AsyncSpeechToTextTranslateStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextTranslateStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextTranslateStreamingFlushSignal] = None,
-        stream_ongoing_speech_results: typing.Optional[SpeechToTextTranslateStreamingStreamOngoingSpeechResults] = None,
-        streaming_ongoing_requests_frame_size: typing.Optional[str] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncSpeechToTextTranslateStreamingSocketClient]:
@@ -201,12 +182,6 @@ class AsyncSpeechToTextTranslateStreamingClient:
         flush_signal : typing.Optional[SpeechToTextTranslateStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription and translation
 
-        stream_ongoing_speech_results : typing.Optional[SpeechToTextTranslateStreamingStreamOngoingSpeechResults]
-            Enable streaming of ongoing speech results during active speech
-
-        streaming_ongoing_requests_frame_size : typing.Optional[str]
-            Frame size for streaming ongoing speech results (1-100)
-
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -231,12 +206,6 @@ 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 stream_ongoing_speech_results is not None:
-            query_params = query_params.add("stream_ongoing_speech_results", stream_ongoing_speech_results)
-        if streaming_ongoing_requests_frame_size is not None:
-            query_params = query_params.add(
-                "streaming_ongoing_requests_frame_size", streaming_ongoing_requests_frame_size
-            )
         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_translate_streaming/raw_client.py

@@ -15,9 +15,6 @@ 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_stream_ongoing_speech_results import (
-    SpeechToTextTranslateStreamingStreamOngoingSpeechResults,
-)
 from .types.speech_to_text_translate_streaming_vad_signals import SpeechToTextTranslateStreamingVadSignals
 
 try:
@@ -40,8 +37,6 @@ class RawSpeechToTextTranslateStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextTranslateStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextTranslateStreamingFlushSignal] = None,
-        stream_ongoing_speech_results: typing.Optional[SpeechToTextTranslateStreamingStreamOngoingSpeechResults] = None,
-        streaming_ongoing_requests_frame_size: typing.Optional[str] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[SpeechToTextTranslateStreamingSocketClient]:
@@ -72,12 +67,6 @@ class RawSpeechToTextTranslateStreamingClient:
         flush_signal : typing.Optional[SpeechToTextTranslateStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription and translation
 
-        stream_ongoing_speech_results : typing.Optional[SpeechToTextTranslateStreamingStreamOngoingSpeechResults]
-            Enable streaming of ongoing speech results during active speech
-
-        streaming_ongoing_requests_frame_size : typing.Optional[str]
-            Frame size for streaming ongoing speech results (1-100)
-
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -102,12 +91,6 @@ class RawSpeechToTextTranslateStreamingClient:
             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 stream_ongoing_speech_results is not None:
-            query_params = query_params.add("stream_ongoing_speech_results", stream_ongoing_speech_results)
-        if streaming_ongoing_requests_frame_size is not None:
-            query_params = query_params.add(
-                "streaming_ongoing_requests_frame_size", streaming_ongoing_requests_frame_size
-            )
         ws_url = ws_url + f"?{query_params}"
         headers = self._client_wrapper.get_headers()
         if api_subscription_key is not None:
@@ -146,8 +129,6 @@ class AsyncRawSpeechToTextTranslateStreamingClient:
         high_vad_sensitivity: typing.Optional[SpeechToTextTranslateStreamingHighVadSensitivity] = None,
         vad_signals: typing.Optional[SpeechToTextTranslateStreamingVadSignals] = None,
         flush_signal: typing.Optional[SpeechToTextTranslateStreamingFlushSignal] = None,
-        stream_ongoing_speech_results: typing.Optional[SpeechToTextTranslateStreamingStreamOngoingSpeechResults] = None,
-        streaming_ongoing_requests_frame_size: typing.Optional[str] = None,
         api_subscription_key: typing.Optional[str] = None,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[AsyncSpeechToTextTranslateStreamingSocketClient]:
@@ -178,12 +159,6 @@ class AsyncRawSpeechToTextTranslateStreamingClient:
         flush_signal : typing.Optional[SpeechToTextTranslateStreamingFlushSignal]
             Signal to flush the audio buffer and finalize transcription and translation
 
-        stream_ongoing_speech_results : typing.Optional[SpeechToTextTranslateStreamingStreamOngoingSpeechResults]
-            Enable streaming of ongoing speech results during active speech
-
-        streaming_ongoing_requests_frame_size : typing.Optional[str]
-            Frame size for streaming ongoing speech results (1-100)
-
         api_subscription_key : typing.Optional[str]
             API subscription key for authentication
 
@@ -208,12 +183,6 @@ class AsyncRawSpeechToTextTranslateStreamingClient:
             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 stream_ongoing_speech_results is not None:
-            query_params = query_params.add("stream_ongoing_speech_results", stream_ongoing_speech_results)
-        if streaming_ongoing_requests_frame_size is not None:
-            query_params = query_params.add(
-                "streaming_ongoing_requests_frame_size", streaming_ongoing_requests_frame_size
-            )
         ws_url = ws_url + f"?{query_params}"
         headers = self._client_wrapper.get_headers()
         if api_subscription_key is not None:

sarvamai/speech_to_text_translate_streaming/types/__init__.py

@@ -5,15 +5,11 @@
 from .speech_to_text_translate_streaming_flush_signal import SpeechToTextTranslateStreamingFlushSignal
 from .speech_to_text_translate_streaming_high_vad_sensitivity import SpeechToTextTranslateStreamingHighVadSensitivity
 from .speech_to_text_translate_streaming_input_audio_codec import SpeechToTextTranslateStreamingInputAudioCodec
-from .speech_to_text_translate_streaming_stream_ongoing_speech_results import (
-    SpeechToTextTranslateStreamingStreamOngoingSpeechResults,
-)
 from .speech_to_text_translate_streaming_vad_signals import SpeechToTextTranslateStreamingVadSignals
 
 __all__ = [
     "SpeechToTextTranslateStreamingFlushSignal",
     "SpeechToTextTranslateStreamingHighVadSensitivity",
     "SpeechToTextTranslateStreamingInputAudioCodec",
-    "SpeechToTextTranslateStreamingStreamOngoingSpeechResults",
     "SpeechToTextTranslateStreamingVadSignals",
 ]

sarvamai/types/__init__.py

@@ -57,7 +57,6 @@ from .language_identification_response import LanguageIdentificationResponse
 from .numerals_format import NumeralsFormat
 from .ping_signal import PingSignal
 from .reasoning_effort import ReasoningEffort
-from .response_speech_state import ResponseSpeechState
 from .response_type import ResponseType
 from .role import Role
 from .sarvam_model_ids import SarvamModelIds
@@ -157,7 +156,6 @@ __all__ = [
     "NumeralsFormat",
     "PingSignal",
     "ReasoningEffort",
-    "ResponseSpeechState",
     "ResponseType",
     "Role",
     "SarvamModelIds",

sarvamai/types/completion_event_flag.py

@@ -1,3 +1,5 @@
 # This file was auto-generated by Fern from our API Definition.
 
-CompletionEventFlag = bool
+import typing
+
+CompletionEventFlag = typing.Union[typing.Literal["true", "false"], typing.Any]

sarvamai/types/speech_to_text_transcription_data.py

@@ -4,7 +4,6 @@ import typing
 
 import pydantic
 from ..core.pydantic_utilities import IS_PYDANTIC_V2, UniversalBaseModel
-from .response_speech_state import ResponseSpeechState
 from .transcription_metrics import TranscriptionMetrics
 
 
@@ -34,11 +33,6 @@ class SpeechToTextTranscriptionData(UniversalBaseModel):
     BCP-47 code of detected language
     """
 
-    response_speech_state: typing.Optional[ResponseSpeechState] = pydantic.Field(default=None)
-    """
-    Current state of speech detection and processing
-    """
-
     metrics: TranscriptionMetrics
 
     if IS_PYDANTIC_V2:

sarvamai/types/speech_to_text_translate_transcription_data.py

@@ -4,7 +4,6 @@ import typing
 
 import pydantic
 from ..core.pydantic_utilities import IS_PYDANTIC_V2, UniversalBaseModel
-from .response_speech_state import ResponseSpeechState
 from .transcription_metrics import TranscriptionMetrics
 
 
@@ -24,11 +23,6 @@ class SpeechToTextTranslateTranscriptionData(UniversalBaseModel):
     BCP-47 code of detected source language (null when language detection is in progress)
     """
 
-    response_speech_state: typing.Optional[ResponseSpeechState] = pydantic.Field(default=None)
-    """
-    Current state of speech detection and processing
-    """
-
     metrics: TranscriptionMetrics
 
     if IS_PYDANTIC_V2:

{sarvamai-0.1.22a3.dist-info → sarvamai-0.1.22a4.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sarvamai
-Version: 0.1.22a3
+Version: 0.1.22a4
 Summary: 
 Requires-Python: >=3.8,<4.0
 Classifier: Intended Audience :: Developers

{sarvamai-0.1.22a3.dist-info → sarvamai-0.1.22a4.dist-info}/RECORD

@@ -1,11 +1,11 @@
-sarvamai/__init__.py,sha256=ST9aZ-zVv6Vh_IrhELC1WSqATPg-Gm7nbrRdK6ZEN_g,11785
+sarvamai/__init__.py,sha256=vI_MiA00kkNnsVNJmIFkf1o_-vLIp8SeNy4xdzjQ_wY,11499
 sarvamai/chat/__init__.py,sha256=_VhToAyIt_5axN6CLJwtxg3-CO7THa_23pbUzqhXJa4,85
 sarvamai/chat/client.py,sha256=xOSj83Gr6Q7eY2qUeATiuXYQqBqWqSCQlIEopK5fKus,11022
 sarvamai/chat/raw_client.py,sha256=A2kRuZcVWlJhyYCD7YKgqNkZEp3cYa1731KhRkhirU0,17885
 sarvamai/client.py,sha256=J30X_os1lPf8Wml0KDFEf6p8VGHhgF_lf3nw1T2D3qo,8207
 sarvamai/core/__init__.py,sha256=YE2CtXeASe1RAbaI39twKWYKCuT4tW5is9HWHhJjR_g,1653
 sarvamai/core/api_error.py,sha256=44vPoTyWN59gonCIZMdzw7M1uspygiLnr3GNFOoVL2Q,614
-sarvamai/core/client_wrapper.py,sha256=METOzaadK7zwtrbukCi36XZ59jVQR35bQGPUzSKmwH8,2570
+sarvamai/core/client_wrapper.py,sha256=4xbcx3NSJ2H-l5jU2jidpdocEhri8wqKQnNFN1Peah0,2570
 sarvamai/core/datetime_utils.py,sha256=nBys2IsYrhPdszxGKCNRPSOCwa-5DWOHG95FB8G9PKo,1047
 sarvamai/core/events.py,sha256=HvKBdSoYcFetk7cgNXb7FxuY-FtY8NtUhZIN7mGVx8U,1159
 sarvamai/core/file.py,sha256=d4NNbX8XvXP32z8KpK2Xovv33nFfruIrpz0QWxlgpZk,2663
@@ -71,12 +71,12 @@ sarvamai/requests/speech_to_text_job_parameters.py,sha256=F3V6XvdUQlWlB4-5XAcRsF
 sarvamai/requests/speech_to_text_response.py,sha256=GS3jNmHDOxqNZ7cvftD62khUMSBIQUu6zEPdCqk8zJk,1041
 sarvamai/requests/speech_to_text_response_data.py,sha256=69fYRdL0tCKpgKQqwzcM4T4Nf_lRxJFh-VCFe_tN964,364
 sarvamai/requests/speech_to_text_streaming_response.py,sha256=cN5tKE9wOWuyBna4wmrf-0LfkOULMpRaJ7qjLuu76V0,348
-sarvamai/requests/speech_to_text_transcription_data.py,sha256=ZbrUy4SBEx8OoeOW1PwTbTq8171Z2z2H1Mh_ZZxrREI,1087
+sarvamai/requests/speech_to_text_transcription_data.py,sha256=Vc65hXDq65d14cP-fDJm151bi7XEKgPItNGt1UL6cOY,877
 sarvamai/requests/speech_to_text_translate_job_parameters.py,sha256=Cco38i8IhX00S2eW3MhLekqUFMS7hZW2AwbpWyCAgpU,990
 sarvamai/requests/speech_to_text_translate_response.py,sha256=xLV2F37PkGR0erRDfTBEPWvywR8eVSL9JbH5a0C9wkY,893
 sarvamai/requests/speech_to_text_translate_response_data.py,sha256=OmjunP9R2xertJKn4fmpyzjDdj1_B_Yh6ZjH1eOlR-Q,407
 sarvamai/requests/speech_to_text_translate_streaming_response.py,sha256=KTjYZ0_oLapuM5Iiq7UwejMsrL1TGgFAW4k5l17TkZs,385
-sarvamai/requests/speech_to_text_translate_transcription_data.py,sha256=xBVSuFtYz-tNiE69OMFpfBnxbEG3ZMl7YMFDWoJyhWU,805
+sarvamai/requests/speech_to_text_translate_transcription_data.py,sha256=oAmW5ihTd301IJYN2u2KrZxB0j3EMacFBfvIhtOSjFI,595
 sarvamai/requests/stop_configuration.py,sha256=Xmp8zyUpnN65pH5A7NqefckB8wk53_BBzOUrgRm2gXs,146
 sarvamai/requests/stt_flush_signal.py,sha256=Gb-SoPPAyVKFVPZKxebLgV4bAv21NjVgvfCl5cqcxrY,360
 sarvamai/requests/task_detail_v_1.py,sha256=2rehl7dSDSgzaw13b9bODamhiN2uB-IK4cOksq8Vmqc,582
@@ -91,32 +91,30 @@ sarvamai/speech_to_text/client.py,sha256=JZ-3ZenTTcJkSAIb7Hkj8zBS4r2TV_jlynl4Ljx
 sarvamai/speech_to_text/raw_client.py,sha256=oCGHyVtVBpXGRSWtAvR_r2j3tEumc9VEDjOYqMVxv5w,26677
 sarvamai/speech_to_text_job/__init__.py,sha256=_VhToAyIt_5axN6CLJwtxg3-CO7THa_23pbUzqhXJa4,85
 sarvamai/speech_to_text_job/client.py,sha256=WSGBJxYcNxl77Zd1X6VVWjg4zshqecXf6WCyhfLXVlI,18007
-sarvamai/speech_to_text_job/job.py,sha256=K8HOmwrYd6l82-MZfWDBmNkZeeERyg9YOihnFfvl-Js,15021
-sarvamai/speech_to_text_job/raw_client.py,sha256=OZTPzMhAn-ckE_xKzfZ9QLsEX5EZVOJS0Pf-PBa19jM,48200
-sarvamai/speech_to_text_streaming/__init__.py,sha256=6xLJUFTGQOnZEeUSV_lL9T2zpQCJ0WP_O-imYFVhrIE,643
-sarvamai/speech_to_text_streaming/client.py,sha256=wGTtFHGxVmpk6F9mpyzgHTcEKlTZChfbG2ytEeb9b3E,13635
-sarvamai/speech_to_text_streaming/raw_client.py,sha256=xlbIKnDrGggXsKubYbpbg5g7hFZ_FHGsozXT2jyxhrM,12816
+sarvamai/speech_to_text_job/job.py,sha256=9AfVSp5nzrl-Cx_1n2AJZqTMzp6Dkz2cvmbdq78fCgM,18751
+sarvamai/speech_to_text_job/raw_client.py,sha256=6MB82mSqAOi92mE8vUeNSTB0wuxLZYRwizt15R6r-wo,49394
+sarvamai/speech_to_text_streaming/__init__.py,sha256=-7nN6AJFryjSvGHVbajYEt-vni6kNDfJUiZJFNl_ao4,535
+sarvamai/speech_to_text_streaming/client.py,sha256=GU3bin0Ea6rOoX-hc9iUStPU0rh6ThezKW-r8d3NgTg,11732
+sarvamai/speech_to_text_streaming/raw_client.py,sha256=lUxjNbdxfirnAcPd4mMhd0EK7c0Z1i9Fa-hHscviFmg,10913
 sarvamai/speech_to_text_streaming/socket_client.py,sha256=P6qXRN0s3UFAp6CP5lkqrW2KPK9me70ZVfWquxLB4wI,7538
-sarvamai/speech_to_text_streaming/types/__init__.py,sha256=cHFJXPoXFFPxkwfFIcXbESlHG9NUFWDUmLQ8bPmCgcQ,926
+sarvamai/speech_to_text_streaming/types/__init__.py,sha256=_G5TSTthsnjGmwdV4fpsybjEWMMTNkh-kWXZjgK5X48,755
 sarvamai/speech_to_text_streaming/types/speech_to_text_streaming_flush_signal.py,sha256=dDJOBlzAjhuiSVqW2RHHY1f6xy0DU_Yoo9UV8-7MjnA,173
 sarvamai/speech_to_text_streaming/types/speech_to_text_streaming_high_vad_sensitivity.py,sha256=OwPwffa8TkLPGMnOTn5S7d-HmV8QmN3B7fHz8I1-VT8,180
 sarvamai/speech_to_text_streaming/types/speech_to_text_streaming_input_audio_codec.py,sha256=yfINVbnoiAvhDskwVkpnFv10Q9SzU104oRJR-n__ugc,584
 sarvamai/speech_to_text_streaming/types/speech_to_text_streaming_language_code.py,sha256=LxgEifmgWTCFZn9U-f-TWKxRPng3a2J26Zt526QrA0Y,267
-sarvamai/speech_to_text_streaming/types/speech_to_text_streaming_stream_ongoing_speech_results.py,sha256=_hc9wuLDsRB6aahsmHqYK0tLoLb6MFZjNo3N7enj8mI,188
 sarvamai/speech_to_text_streaming/types/speech_to_text_streaming_vad_signals.py,sha256=8wiFOB7WDMbYCcMTYgNFJaIjEytYeXpJLwr_O_mH0TI,172
 sarvamai/speech_to_text_translate_job/__init__.py,sha256=_VhToAyIt_5axN6CLJwtxg3-CO7THa_23pbUzqhXJa4,85
 sarvamai/speech_to_text_translate_job/client.py,sha256=xu8kYtCESDB7LzL8YKBUq5qhTPMIl3_H3XD2L_7y4UU,18969
-sarvamai/speech_to_text_translate_job/job.py,sha256=DU4k3eB28V8N16M_QEchakVng4IOul6_Qrdn3FumgHA,15208
-sarvamai/speech_to_text_translate_job/raw_client.py,sha256=dAitbu2B9afPK6iT9zNjUJnE5BIr5-lrAlwrfwFxdkU,49507
-sarvamai/speech_to_text_translate_streaming/__init__.py,sha256=KmUM04hMxqYEG6I7rDFmlCBIYgHDIG8-A9d8bKLImg4,653
-sarvamai/speech_to_text_translate_streaming/client.py,sha256=8Gi22KsMlhn4QKyNeH59vGGF3QpN5RwMoT2G7BjzP2E,13617
-sarvamai/speech_to_text_translate_streaming/raw_client.py,sha256=jrWDHke_HApWIxBeGoyFLwEwYnYtx-Mj2M5UuGp-ivk,12726
+sarvamai/speech_to_text_translate_job/job.py,sha256=tL1Zemsogb_AK9wqZwN4ooPaN176sFKduTH9g87y-WU,18938
+sarvamai/speech_to_text_translate_job/raw_client.py,sha256=Emx14cRiAZXg1PqZkoJbDOKwyDmOgwxWlqPkAPZ9GPU,50797
+sarvamai/speech_to_text_translate_streaming/__init__.py,sha256=s6HPwrkABpkhDSsd_t6pVRiWfY4MfVE0lVj9b4V_fx4,527
+sarvamai/speech_to_text_translate_streaming/client.py,sha256=jXfucb5A5uyKgbAB_tJDYCn86Myv8k8Cs-BIBqaZGHo,11659
+sarvamai/speech_to_text_translate_streaming/raw_client.py,sha256=1TKO5mrqI234wGWPlcyKS-9ObLuY6XRcgYExxv1CJhY,10768
 sarvamai/speech_to_text_translate_streaming/socket_client.py,sha256=ipEPSj5eHAyDpuEXfaP7JJL1rXJXGEo-IB888ReAFKs,8901
-sarvamai/speech_to_text_translate_streaming/types/__init__.py,sha256=L5lAQwKmB57xee93CkgZx3DWWUNcTV-gVNCsEmmbfcs,948
+sarvamai/speech_to_text_translate_streaming/types/__init__.py,sha256=nsKmvwkhcPekF9kcStDhTDilALFf2jT-wfCn25KVe7U,740
 sarvamai/speech_to_text_translate_streaming/types/speech_to_text_translate_streaming_flush_signal.py,sha256=jkjvCGJ1pFKi3AOTkwMW-lo18WGgrgAhMpoe5P0AMzA,182
 sarvamai/speech_to_text_translate_streaming/types/speech_to_text_translate_streaming_high_vad_sensitivity.py,sha256=r6MvTlkM0VEpb4dpnMHtINOZ-gYc22o0Fx_Xce2rjvo,189
 sarvamai/speech_to_text_translate_streaming/types/speech_to_text_translate_streaming_input_audio_codec.py,sha256=IwO6IF9BQU5MpGWs7m-w0dV8yp__7s4y3L4hHDErBWM,593
-sarvamai/speech_to_text_translate_streaming/types/speech_to_text_translate_streaming_stream_ongoing_speech_results.py,sha256=OECioY4SsX5UKsxbE_HIIMZYQEzRg49Z0RhPSRdTfCc,197
 sarvamai/speech_to_text_translate_streaming/types/speech_to_text_translate_streaming_vad_signals.py,sha256=EV3xd9qyKMnMvA9rO-qFDDIac4b84roBu7n-maaPxG8,181
 sarvamai/text/__init__.py,sha256=_VhToAyIt_5axN6CLJwtxg3-CO7THa_23pbUzqhXJa4,85
 sarvamai/text/client.py,sha256=2kA0Gxfi-r52zMQdqRRD811014alzlHB_FANkp3Kn_c,30595
@@ -130,7 +128,7 @@ sarvamai/text_to_speech_streaming/raw_client.py,sha256=E-4cVcgYA1MQRq2YkLoGt1OU3
 sarvamai/text_to_speech_streaming/socket_client.py,sha256=aun1nyc5ZYziXP0FmyvkCB0WKa_VdjA0XHYI2-Z_gCo,13973
 sarvamai/text_to_speech_streaming/types/__init__.py,sha256=fx2VJtnbh4RFUETiV2aPGslEQ0lq_us4UjDbQr_2J6I,242
 sarvamai/text_to_speech_streaming/types/text_to_speech_streaming_send_completion_event.py,sha256=cZWm6cAwNwPGF8ZADtnRev_AsgM_xj5ypg2oHHQfgZI,181
-sarvamai/types/__init__.py,sha256=DvAsZu5BYk6rSQSZ02cl1Prc0tGgo0p_9L9uJjzpwWI,8416
+sarvamai/types/__init__.py,sha256=hBtXC7VLxZdrBEPs7-Ym6c3fYlJJS1VxHbfb4J8CcEM,8334
 sarvamai/types/audio_data.py,sha256=g-sQcFXiV0uFAsUJ_TjM3OSnPoCUmNkMORRJHOl152k,1545
 sarvamai/types/audio_data_input_audio_codec.py,sha256=Xb3UEr2wTniW0z8ND4iV051YHhgFazIFOd7q7UGH3nY,572
 sarvamai/types/audio_message.py,sha256=sB4EgkWkWJzipYXobkmM9AYZTTZtCpg_ySKssUeznUE,560
@@ -145,7 +143,7 @@ sarvamai/types/chat_completion_request_system_message.py,sha256=E7YhTk1zr4u7dj_y
 sarvamai/types/chat_completion_request_user_message.py,sha256=J3WhlrfOfCCe7ugmJIfP_L9st3OFtXkIjZTSuR8O9nQ,615
 sarvamai/types/chat_completion_response_message.py,sha256=wz935eBnCkSIl0I0qMxBuH4vAUCso1aHDGReMW1VHGE,744
 sarvamai/types/choice.py,sha256=uXBCsjWP9VK3XWQWZUeI4EnU10w0G9nAfKn2tJZvxko,1244
-sarvamai/types/completion_event_flag.py,sha256=TzDOwE5GrC2uiVpnc9vV8y7n4uT_cr_G5FR011h_zw4,92
+sarvamai/types/completion_event_flag.py,sha256=HdvjxXha9Ux5KS_Lfp7Q5eaX62eIk4bXcuJgfhGFXf4,160
 sarvamai/types/completion_usage.py,sha256=xYQGlQUbKqsksuV73H-1ajjfT5M7w47eLfdWXSlrI5M,843
 sarvamai/types/config_message.py,sha256=xLD2wZcXejYrmREMd-cn38da4hKfsNPKRtyAGCW0Zcg,779
 sarvamai/types/configure_connection.py,sha256=SnSNk02gQqP8e4VB4y88jjeFQ4ClpImjGLn2ANI8cZ4,1058
@@ -181,7 +179,6 @@ sarvamai/types/language_identification_response.py,sha256=jG4ZQ6KQHCiEDqC51OniOw
 sarvamai/types/numerals_format.py,sha256=xg3lYiHcnzyFwuwRcaIteJLH_Pz6pJ9n9kTlYPEnCBU,165
 sarvamai/types/ping_signal.py,sha256=cE53FRIXlc8bSo18z6jlAnOh6DhZEMX36huWEX6X3-A,695
 sarvamai/types/reasoning_effort.py,sha256=_TBLn3rQgzJAdnKqV2g0PETbrSBZl0fPLfQ5ZE9H4Pc,164
-sarvamai/types/response_speech_state.py,sha256=SYwVP-TD2nhZ2kAXw0NzuVtkRgMjbLsuXRfKHPGxhK0,231
 sarvamai/types/response_type.py,sha256=yyk0QTIQlNa9W0Uoj_5_ey_Q3Bu8Jij5GkgR0Rt_WnU,163
 sarvamai/types/role.py,sha256=3eY01zZQKB8BSD4cFDeVjz-o2qnHJKz1vnToLqbExxs,115
 sarvamai/types/sarvam_model_ids.py,sha256=iYBMglf31KQ1iUZeAMQ-2PP9NDcyHRG7goz7O9VG8qg,124
@@ -194,14 +191,14 @@ sarvamai/types/speech_to_text_model.py,sha256=DSnGHPpIT-OyRv1bpy1xqEoLw90MTKyDgc
 sarvamai/types/speech_to_text_response.py,sha256=iWRGEJeHUFIOxeEhoCQu68njeA6lcqXbT2czV-O8Wx0,1438
 sarvamai/types/speech_to_text_response_data.py,sha256=gbxZTBSjbN3ZIa10K6tWPYtymcpnQTFIaUnXkOmsmD4,322
 sarvamai/types/speech_to_text_streaming_response.py,sha256=z6tVAHbVK9lC3w3lac__LEUfO8AAzEilkeGlaLskTtc,687
-sarvamai/types/speech_to_text_transcription_data.py,sha256=9xJljx_Uhhr0_DQZADPHbvEYmVCNwo9XHy7_QFPVqgY,1517
+sarvamai/types/speech_to_text_transcription_data.py,sha256=EqwPAPSi98PwARaTj-ufzFUSHyN-NPoPla5vi_KERrU,1297
 sarvamai/types/speech_to_text_translate_job_parameters.py,sha256=fvfcyzIyT0DtcRYePDvglHH-wAhGbsi3H5G4i5nuWT8,1409
 sarvamai/types/speech_to_text_translate_language.py,sha256=yikNM-roIumVG-eqBWss93wLGudZdLPwd0i3VcXH5zo,263
 sarvamai/types/speech_to_text_translate_model.py,sha256=CVSz6gJBY82GhhEuWSdzRLJW9XTsAgweRnKd1tN6mXo,139
 sarvamai/types/speech_to_text_translate_response.py,sha256=Z5Na7IQW2ok3TP21xd-jKkwioplEKfonNIMhoJQKkVw,1278
 sarvamai/types/speech_to_text_translate_response_data.py,sha256=_NlLVp7oQU3em_4E47QVbIP9nromPE07Z9HtMpY1lrU,359
 sarvamai/types/speech_to_text_translate_streaming_response.py,sha256=J6h3AGdAJxpODFs30bR-e6OaWKa__oVhwv_TrbPSO98,724
-sarvamai/types/speech_to_text_translate_transcription_data.py,sha256=SPdYTeD1H4WQiExfs-jETyTwXblrqlsthTjUHqZEWxw,1216
+sarvamai/types/speech_to_text_translate_transcription_data.py,sha256=-cZZm21um6erOzx18nAgBYKA3Qc3YzNADUCnfx9mD-k,996
 sarvamai/types/spoken_form_numerals_format.py,sha256=soBly93wMkazIcp2GDM0Mf1MjY140Pe24hBlwNoWge0,169
 sarvamai/types/stop_configuration.py,sha256=yA_q4s4BIrbl3FotZpg4ZcyL10C7gVI0s2dqvH32BNw,136
 sarvamai/types/storage_container_type.py,sha256=DZXDiDj74lMmUq6jaZfIMW1zMXgoVdY6rs_FcyB9OGk,184
@@ -227,6 +224,6 @@ sarvamai/types/transliterate_mode.py,sha256=1jSEMlGcoLkWuk12TgoOpSgwifa4rThGKZ1h
 sarvamai/types/transliterate_source_language.py,sha256=bSY9wJszF0sg-Cgg6F-YcWC8ly1mIlj9rqa15-jBtx8,283
 sarvamai/types/transliteration_response.py,sha256=yt-lzTbDeJ_ZL4I8kQa6oESxA9ebeJJY7LfFHpdEsmM,815
 sarvamai/version.py,sha256=Qkp3Ee9YH-O9RTix90e0i7iNrFAGN-QDt2AFwGA4n8k,75
-sarvamai-0.1.22a3.dist-info/METADATA,sha256=Y_DbEm9yqps7R3sCQuZrxb6E73ww6uGni-biEMYFurQ,26753
-sarvamai-0.1.22a3.dist-info/WHEEL,sha256=Zb28QaM1gQi8f4VCBhsUklF61CTlNYfs9YAZn-TOGFk,88
-sarvamai-0.1.22a3.dist-info/RECORD,,
+sarvamai-0.1.22a4.dist-info/METADATA,sha256=UrGlg-Q0Vo4UkotwZRRtKtvLu8vwJNYxGAbkxBS87_I,26753
+sarvamai-0.1.22a4.dist-info/WHEEL,sha256=Zb28QaM1gQi8f4VCBhsUklF61CTlNYfs9YAZn-TOGFk,88
+sarvamai-0.1.22a4.dist-info/RECORD,,

sarvamai/speech_to_text_streaming/types/speech_to_text_streaming_stream_ongoing_speech_results.py

@@ -1,5 +0,0 @@
-# This file was auto-generated by Fern from our API Definition.
-
-import typing
-
-SpeechToTextStreamingStreamOngoingSpeechResults = typing.Union[typing.Literal["true", "false"], typing.Any]

sarvamai/speech_to_text_translate_streaming/types/speech_to_text_translate_streaming_stream_ongoing_speech_results.py

@@ -1,5 +0,0 @@
-# This file was auto-generated by Fern from our API Definition.
-
-import typing
-
-SpeechToTextTranslateStreamingStreamOngoingSpeechResults = typing.Union[typing.Literal["true", "false"], typing.Any]

sarvamai/types/response_speech_state.py

@@ -1,7 +0,0 @@
-# This file was auto-generated by Fern from our API Definition.
-
-import typing
-
-ResponseSpeechState = typing.Union[
-    typing.Literal["START_SPEECH_TENTATIVE", "VAD_MISFIRE", "START_SPEECH", "END_SPEECH", "ONGOING"], typing.Any
-]