sarvamai 0.1.22a3__py3-none-any.whl → 0.1.23a1__py3-none-any.whl

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
 
 
@@ -18,12 +17,12 @@ class SpeechToTextTranscriptionDataParams(typing_extensions.TypedDict):
     Transcript of the provided speech in original language
     """
 
-    timestamps: typing_extensions.NotRequired[typing.Dict[str, typing.Optional[typing.Any]]]
+    timestamps: typing_extensions.NotRequired[typing.Dict[str, typing.Any]]
     """
     Timestamp information (if available)
     """
 
-    diarized_transcript: typing_extensions.NotRequired[typing.Dict[str, typing.Optional[typing.Any]]]
+    diarized_transcript: typing_extensions.NotRequired[typing.Dict[str, typing.Any]]
     """
     Diarized transcript of the provided speech
     """
@@ -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/raw_client.py

@@ -1,5 +1,6 @@
 # This file was auto-generated by Fern from our API Definition.
 
+import json
 import typing
 from json.decoder import JSONDecodeError
 
@@ -7,6 +8,7 @@ from .. import core
 from ..core.api_error import ApiError
 from ..core.client_wrapper import AsyncClientWrapper, SyncClientWrapper
 from ..core.http_response import AsyncHttpResponse, HttpResponse
+from ..core.jsonable_encoder import jsonable_encoder
 from ..core.pydantic_utilities import parse_obj_as
 from ..core.request_options import RequestOptions
 from ..errors.bad_request_error import BadRequestError
@@ -84,7 +86,7 @@ class RawSpeechToTextClient:
             base_url=self._client_wrapper.get_environment().base,
             method="POST",
             data={
-                "model": model,
+                "model": json.dumps(jsonable_encoder(model)),
                 "language_code": language_code,
                 "input_audio_codec": input_audio_codec,
             },
@@ -109,9 +111,9 @@ class RawSpeechToTextClient:
                 raise BadRequestError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -120,9 +122,9 @@ class RawSpeechToTextClient:
                 raise ForbiddenError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -131,9 +133,9 @@ class RawSpeechToTextClient:
                 raise UnprocessableEntityError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -142,9 +144,9 @@ class RawSpeechToTextClient:
                 raise TooManyRequestsError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -153,9 +155,9 @@ class RawSpeechToTextClient:
                 raise InternalServerError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -164,9 +166,9 @@ class RawSpeechToTextClient:
                 raise ServiceUnavailableError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -228,7 +230,7 @@ class RawSpeechToTextClient:
             method="POST",
             data={
                 "prompt": prompt,
-                "model": model,
+                "model": json.dumps(jsonable_encoder(model)),
                 "input_audio_codec": input_audio_codec,
             },
             files={
@@ -252,9 +254,9 @@ class RawSpeechToTextClient:
                 raise BadRequestError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -263,9 +265,9 @@ class RawSpeechToTextClient:
                 raise ForbiddenError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -274,9 +276,9 @@ class RawSpeechToTextClient:
                 raise UnprocessableEntityError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -285,9 +287,9 @@ class RawSpeechToTextClient:
                 raise TooManyRequestsError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -296,9 +298,9 @@ class RawSpeechToTextClient:
                 raise InternalServerError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -307,9 +309,9 @@ class RawSpeechToTextClient:
                 raise ServiceUnavailableError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -378,7 +380,7 @@ class AsyncRawSpeechToTextClient:
             base_url=self._client_wrapper.get_environment().base,
             method="POST",
             data={
-                "model": model,
+                "model": json.dumps(jsonable_encoder(model)),
                 "language_code": language_code,
                 "input_audio_codec": input_audio_codec,
             },
@@ -403,9 +405,9 @@ class AsyncRawSpeechToTextClient:
                 raise BadRequestError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -414,9 +416,9 @@ class AsyncRawSpeechToTextClient:
                 raise ForbiddenError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -425,9 +427,9 @@ class AsyncRawSpeechToTextClient:
                 raise UnprocessableEntityError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -436,9 +438,9 @@ class AsyncRawSpeechToTextClient:
                 raise TooManyRequestsError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -447,9 +449,9 @@ class AsyncRawSpeechToTextClient:
                 raise InternalServerError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -458,9 +460,9 @@ class AsyncRawSpeechToTextClient:
                 raise ServiceUnavailableError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -522,7 +524,7 @@ class AsyncRawSpeechToTextClient:
             method="POST",
             data={
                 "prompt": prompt,
-                "model": model,
+                "model": json.dumps(jsonable_encoder(model)),
                 "input_audio_codec": input_audio_codec,
             },
             files={
@@ -546,9 +548,9 @@ class AsyncRawSpeechToTextClient:
                 raise BadRequestError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -557,9 +559,9 @@ class AsyncRawSpeechToTextClient:
                 raise ForbiddenError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -568,9 +570,9 @@ class AsyncRawSpeechToTextClient:
                 raise UnprocessableEntityError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -579,9 +581,9 @@ class AsyncRawSpeechToTextClient:
                 raise TooManyRequestsError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -590,9 +592,9 @@ class AsyncRawSpeechToTextClient:
                 raise InternalServerError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),
@@ -601,9 +603,9 @@ class AsyncRawSpeechToTextClient:
                 raise ServiceUnavailableError(
                     headers=dict(_response.headers),
                     body=typing.cast(
-                        typing.Optional[typing.Any],
+                        typing.Any,
                         parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
+                            type_=typing.Any,  # type: ignore
                             object_=_response.json(),
                         ),
                     ),

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.