sarvamai 0.1.22a3__tar.gz → 0.1.22a4__tar.gz

{sarvamai-0.1.22a3 → sarvamai-0.1.22a4}/PKG-INFO

@@ -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 → sarvamai-0.1.22a4}/pyproject.toml

@@ -3,7 +3,7 @@ name = "sarvamai"
 
 [tool.poetry]
 name = "sarvamai"
-version = "0.1.22a3"
+version = "0.1.22a4"
 description = ""
 readme = "README.md"
 authors = []

{sarvamai-0.1.22a3 → sarvamai-0.1.22a4}/src/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-0.1.22a3 → sarvamai-0.1.22a4}/src/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-0.1.22a3 → sarvamai-0.1.22a4}/src/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-0.1.22a3 → sarvamai-0.1.22a4}/src/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-0.1.22a3 → sarvamai-0.1.22a4}/src/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-0.1.22a3 → sarvamai-0.1.22a4}/src/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-0.1.22a3 → sarvamai-0.1.22a4}/src/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-0.1.22a3 → sarvamai-0.1.22a4}/src/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-0.1.22a3 → sarvamai-0.1.22a4}/src/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-0.1.22a3 → sarvamai-0.1.22a4}/src/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-0.1.22a3 → sarvamai-0.1.22a4}/src/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.