sarvamai 0.1.11a3__tar.gz → 0.1.11a5__tar.gz

{sarvamai-0.1.11a3 → sarvamai-0.1.11a5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sarvamai
-Version: 0.1.11a3
+Version: 0.1.11a5
 Summary: 
 Requires-Python: >=3.8,<4.0
 Classifier: Intended Audience :: Developers

{sarvamai-0.1.11a3 → sarvamai-0.1.11a5}/pyproject.toml

@@ -3,7 +3,7 @@ name = "sarvamai"
 
 [tool.poetry]
 name = "sarvamai"
-version = "0.1.11a3"
+version = "0.1.11a5"
 description = ""
 readme = "README.md"
 authors = []

{sarvamai-0.1.11a3 → sarvamai-0.1.11a5}/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/0.1.11a3",
+            "User-Agent": "sarvamai/0.1.11a5",
             "X-Fern-Language": "Python",
             "X-Fern-SDK-Name": "sarvamai",
-            "X-Fern-SDK-Version": "0.1.11a3",
+            "X-Fern-SDK-Version": "0.1.11a5",
             **(self.get_custom_headers() or {}),
         }
         headers["api-subscription-key"] = self.api_subscription_key

{sarvamai-0.1.11a3 → sarvamai-0.1.11a5}/src/sarvamai/speech_to_text_job/client.py

@@ -234,12 +234,53 @@ class SpeechToTextJobClient:
 
     def create_job(
         self,
-        job_parameters: SpeechToTextJobParametersParams,
+        language_code: typing.Optional[str] = None,
+        model: typing.Optional[str] = "saarika:v2.5",
+        num_speakers: typing.Optional[int] = None,
+        with_diarization: typing.Optional[bool] = False,
+        with_timestamps: typing.Optional[bool] = False,
         callback: typing.Optional[BulkJobCallbackParams] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> SpeechToTextJob:
+        """
+        Create a new Speech-to-Text bulk job.
+
+        Parameters
+        ----------
+        language_code : typing.Optional[str], default=None
+            The language code of the input audio (e.g., "en", "hi").
+
+        model : typing.Optional[str], default="saarika:v2.5"
+            The model to use for transcription.
+
+        num_speakers : typing.Optional[int], default=None
+            The number of distinct speakers in the audio, if known.
+
+        with_diarization : typing.Optional[bool], default=False
+            Whether to enable speaker diarization (distinguishing who said what).
+
+        with_timestamps : typing.Optional[bool], default=False
+            Whether to include word-level timestamps in the transcription output.
+
+        callback : typing.Optional[BulkJobCallbackParams], default=OMIT
+            Optional callback configuration to receive job completion events.
+
+        request_options : typing.Optional[RequestOptions], default=None
+            Request-specific configuration.
+
+        Returns
+        -------
+        SpeechToTextJob
+            A handle to the newly created Speech-to-Text job.
+        """
         response = self.initialise(
-            job_parameters=job_parameters,
+            job_parameters=SpeechToTextJobParametersParams(
+                language_code=language_code,
+                model=model,
+                num_speakers=num_speakers,  # type: ignore[typeddict-item]
+                with_diarization=with_diarization,  # type: ignore[typeddict-item]
+                with_timestamps=with_timestamps,  # type: ignore[typeddict-item]
+            ),
             callback=callback,
             request_options=request_options,
         )
@@ -247,7 +288,17 @@ class SpeechToTextJobClient:
 
     def get_job(self, job_id: str) -> SpeechToTextJob:
         """
-        Return a job handle for an existing Speech-to-Text job.
+        Get an existing Speech-to-Text job handle by job ID.
+
+        Parameters
+        ----------
+        job_id : str
+            The job ID of the previously created Speech-to-Text job.
+
+        Returns
+        -------
+        SpeechToTextJob
+            A job handle which can be used to check status or retrieve results.
         """
         return SpeechToTextJob(job_id=job_id, client=self)
 
@@ -511,12 +562,53 @@ class AsyncSpeechToTextJobClient:
 
     async def create_job(
         self,
-        job_parameters: SpeechToTextJobParametersParams,
+        language_code: typing.Optional[str] = None,
+        model: typing.Optional[str] = "saarika:v2.5",
+        num_speakers: typing.Optional[int] = None,
+        with_diarization: typing.Optional[bool] = False,
+        with_timestamps: typing.Optional[bool] = False,
         callback: typing.Optional[BulkJobCallbackParams] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> "AsyncSpeechToTextJob":
+        """
+        Create a new Speech-to-Text bulk job.
+
+        Parameters
+        ----------
+        language_code : typing.Optional[str], default=None
+            The language code of the input audio (e.g., "en", "hi").
+
+        model : typing.Optional[str], default="saarika:v2.5"
+            The model to use for transcription.
+
+        num_speakers : typing.Optional[int], default=None
+            The number of distinct speakers in the audio, if known.
+
+        with_diarization : typing.Optional[bool], default=False
+            Whether to enable speaker diarization (distinguishing who said what).
+
+        with_timestamps : typing.Optional[bool], default=False
+            Whether to include word-level timestamps in the transcription output.
+
+        callback : typing.Optional[BulkJobCallbackParams], default=OMIT
+            Optional callback configuration to receive job completion events.
+
+        request_options : typing.Optional[RequestOptions], default=None
+            Request-specific configuration.
+
+        Returns
+        -------
+        AsyncSpeechToTextJob
+            A handle to the newly created job.
+        """
         response = await self.initialise(
-            job_parameters=job_parameters,
+            job_parameters=SpeechToTextJobParametersParams(
+                language_code=language_code,
+                model=model,
+                num_speakers=num_speakers,  # type: ignore[typeddict-item]
+                with_diarization=with_diarization,  # type: ignore[typeddict-item]
+                with_timestamps=with_timestamps,  # type: ignore[typeddict-item]
+            ),
             callback=callback,
             request_options=request_options,
         )
@@ -524,6 +616,16 @@ class AsyncSpeechToTextJobClient:
 
     async def get_job(self, job_id: str) -> "AsyncSpeechToTextJob":
         """
-        Return a job handle for an existing Speech-to-Text job.
+        Get an existing Speech-to-Text job handle by job ID.
+
+        Parameters
+        ----------
+        job_id : str
+            The job ID of the previously created speech-to-text job.
+
+        Returns
+        -------
+        AsyncSpeechToTextJob
+            A job handle which can be used to check status or retrieve results.
         """
         return AsyncSpeechToTextJob(job_id=job_id, client=self)

{sarvamai-0.1.11a3 → sarvamai-0.1.11a5}/src/sarvamai/speech_to_text_job/job.py

@@ -1,4 +1,5 @@
 import asyncio
+import mimetypes
 import os
 import time
 import typing
@@ -66,12 +67,15 @@ class AsyncSpeechToTextJob:
                 file_name = os.path.basename(path)
                 url = upload_links.upload_urls[file_name].file_url
                 with open(path, "rb") as f:
+                    content_type, _ = mimetypes.guess_type(path)
+                    if content_type is None:
+                        content_type = "audio/wav"
                     response = await session.put(
                         url,
                         content=f.read(),
                         headers={
                             "x-ms-blob-type": "BlockBlob",
-                            "Content-Type": "audio/wav",
+                            "Content-Type": content_type,
                         },
                     )
                 if response.status_code != 201:

{sarvamai-0.1.11a3 → sarvamai-0.1.11a5}/src/sarvamai/speech_to_text_translate_job/client.py

@@ -5,12 +5,18 @@ import typing
 from ..core.client_wrapper import AsyncClientWrapper, SyncClientWrapper
 from ..core.request_options import RequestOptions
 from ..requests.bulk_job_callback import BulkJobCallbackParams
-from ..requests.speech_to_text_translate_job_parameters import SpeechToTextTranslateJobParametersParams
+from ..requests.speech_to_text_translate_job_parameters import (
+    SpeechToTextTranslateJobParametersParams,
+)
 from ..types.bulk_job_init_response_v_1 import BulkJobInitResponseV1
 from ..types.files_download_response import FilesDownloadResponse
 from ..types.files_upload_response import FilesUploadResponse
 from ..types.job_status_v_1 import JobStatusV1
-from .raw_client import AsyncRawSpeechToTextTranslateJobClient, RawSpeechToTextTranslateJobClient
+from .job import AsyncSpeechToTextTranslateJob, SpeechToTextTranslateJob
+from .raw_client import (
+    AsyncRawSpeechToTextTranslateJobClient,
+    RawSpeechToTextTranslateJobClient,
+)
 
 # this is used as the default value for optional parameters
 OMIT = typing.cast(typing.Any, ...)
@@ -18,7 +24,9 @@ OMIT = typing.cast(typing.Any, ...)
 
 class SpeechToTextTranslateJobClient:
     def __init__(self, *, client_wrapper: SyncClientWrapper):
-        self._raw_client = RawSpeechToTextTranslateJobClient(client_wrapper=client_wrapper)
+        self._raw_client = RawSpeechToTextTranslateJobClient(
+            client_wrapper=client_wrapper
+        )
 
     @property
     def with_raw_response(self) -> RawSpeechToTextTranslateJobClient:
@@ -72,11 +80,16 @@ class SpeechToTextTranslateJobClient:
         )
         """
         _response = self._raw_client.initialise(
-            job_parameters=job_parameters, ptu_id=ptu_id, callback=callback, request_options=request_options
+            job_parameters=job_parameters,
+            ptu_id=ptu_id,
+            callback=callback,
+            request_options=request_options,
         )
         return _response.data
 
-    def get_status(self, job_id: str, *, request_options: typing.Optional[RequestOptions] = None) -> JobStatusV1:
+    def get_status(
+        self, job_id: str, *, request_options: typing.Optional[RequestOptions] = None
+    ) -> JobStatusV1:
         """
         Get the status of a speech to text translate bulk job V1
 
@@ -143,7 +156,9 @@ class SpeechToTextTranslateJobClient:
             job_id="job_id",
         )
         """
-        _response = self._raw_client.start(job_id, ptu_id=ptu_id, request_options=request_options)
+        _response = self._raw_client.start(
+            job_id, ptu_id=ptu_id, request_options=request_options
+        )
         return _response.data
 
     def get_upload_links(
@@ -234,10 +249,77 @@ class SpeechToTextTranslateJobClient:
         )
         return _response.data
 
+    def create_job(
+        self,
+        prompt: typing.Optional[str] = None,
+        model: typing.Optional[str] = "saaras:v2.5",
+        num_speakers: typing.Optional[int] = None,
+        with_diarization: typing.Optional[bool] = False,
+        callback: typing.Optional[BulkJobCallbackParams] = OMIT,
+        request_options: typing.Optional[RequestOptions] = None,
+    ) -> SpeechToTextTranslateJob:
+        """
+        Create a new Speech-to-Text-Translate bulk job.
+
+        Parameters
+        ----------
+        prompt : typing.Optional[str], default=None
+            An optional prompt to guide the transcription and translation model.
+
+        model : typing.Optional[str], default="saaras:v2.5"
+            The model to use for speech-to-text translation.
+
+        num_speakers : typing.Optional[int], default=None
+            The number of distinct speakers in the input audio, if known.
+
+        with_diarization : typing.Optional[bool], default=False
+            Whether to enable speaker diarization (i.e., distinguishing who is speaking).
+
+        callback : typing.Optional[BulkJobCallbackParams], default=OMIT
+            Optional callback configuration to receive job completion events via webhook.
+
+        request_options : typing.Optional[RequestOptions], default=None
+            Optional configuration for request timeout, retries, etc.
+
+        Returns
+        -------
+        SpeechToTextTranslateJob
+            A handle to the newly created Speech-to-Text-Translate job.
+        """
+        response = self.initialise(
+            job_parameters=SpeechToTextTranslateJobParametersParams(
+                prompt=prompt,  # type: ignore[typeddict-item]
+                model=model,
+                with_diarization=with_diarization,  # type: ignore[typeddict-item]
+                num_speakers=num_speakers,  # type: ignore[typeddict-item]
+            ),
+            callback=callback,
+            request_options=request_options,
+        )
+        return SpeechToTextTranslateJob(job_id=response.job_id, client=self)
+
+    def get_job(self, job_id: str) -> SpeechToTextTranslateJob:
+        """
+        Get an existing Speech-to-Text-Translate job handle by job ID.
+
+        Parameters
+        ----------
+        job_id : str
+            The job ID of the previously created Speech-to-Text-Translate job.
+
+        Returns
+        -------
+        SpeechToTextTranslateJob
+            A job handle which can be used to check status or retrieve results.
+        """
+        return SpeechToTextTranslateJob(job_id=job_id, client=self)
+
 
 class AsyncSpeechToTextTranslateJobClient:
     def __init__(self, *, client_wrapper: AsyncClientWrapper):
-        self._raw_client = AsyncRawSpeechToTextTranslateJobClient(client_wrapper=client_wrapper)
+        self._raw_client = AsyncRawSpeechToTextTranslateJobClient(
+            client_wrapper=client_wrapper
+        )
 
     @property
     def with_raw_response(self) -> AsyncRawSpeechToTextTranslateJobClient:
@@ -299,11 +381,16 @@ class AsyncSpeechToTextTranslateJobClient:
         asyncio.run(main())
         """
         _response = await self._raw_client.initialise(
-            job_parameters=job_parameters, ptu_id=ptu_id, callback=callback, request_options=request_options
+            job_parameters=job_parameters,
+            ptu_id=ptu_id,
+            callback=callback,
+            request_options=request_options,
         )
         return _response.data
 
-    async def get_status(self, job_id: str, *, request_options: typing.Optional[RequestOptions] = None) -> JobStatusV1:
+    async def get_status(
+        self, job_id: str, *, request_options: typing.Optional[RequestOptions] = None
+    ) -> JobStatusV1:
         """
         Get the status of a speech to text translate bulk job V1
 
@@ -339,7 +426,9 @@ class AsyncSpeechToTextTranslateJobClient:
 
         asyncio.run(main())
         """
-        _response = await self._raw_client.get_status(job_id, request_options=request_options)
+        _response = await self._raw_client.get_status(
+            job_id, request_options=request_options
+        )
         return _response.data
 
     async def start(
@@ -386,7 +475,9 @@ class AsyncSpeechToTextTranslateJobClient:
 
         asyncio.run(main())
         """
-        _response = await self._raw_client.start(job_id, ptu_id=ptu_id, request_options=request_options)
+        _response = await self._raw_client.start(
+            job_id, ptu_id=ptu_id, request_options=request_options
+        )
         return _response.data
 
     async def get_upload_links(
@@ -492,3 +583,68 @@ class AsyncSpeechToTextTranslateJobClient:
             job_id=job_id, files=files, ptu_id=ptu_id, request_options=request_options
         )
         return _response.data
+
+    async def create_job(
+        self,
+        prompt: typing.Optional[str] = None,
+        model: typing.Optional[str] = "saaras:v2.5",
+        num_speakers: typing.Optional[int] = None,
+        with_diarization: typing.Optional[bool] = False,
+        callback: typing.Optional[BulkJobCallbackParams] = OMIT,
+        request_options: typing.Optional[RequestOptions] = None,
+    ) -> "AsyncSpeechToTextTranslateJob":
+        """
+        Create a new Speech-to-Text-Translate bulk job.
+
+        Parameters
+        ----------
+        prompt : typing.Optional[str], default=None
+            An optional prompt to guide the transcription and translation model.
+
+        model : typing.Optional[str], default="saaras:v2.5"
+            The model to use for speech-to-text translation.
+
+        num_speakers : typing.Optional[int], default=None
+            The number of distinct speakers in the input audio, if known.
+
+        with_diarization : typing.Optional[bool], default=False
+            Whether to enable speaker diarization (i.e., distinguishing who is speaking).
+
+        callback : typing.Optional[BulkJobCallbackParams], default=OMIT
+            Optional callback configuration to receive job completion events via webhook.
+
+        request_options : typing.Optional[RequestOptions], default=None
+            Optional configuration for request timeout, retries, etc.
+
+        Returns
+        -------
+        AsyncSpeechToTextTranslateJob
+            A handle to the newly created job.
+        """
+        response = await self.initialise(
+            job_parameters=SpeechToTextTranslateJobParametersParams(
+                prompt=prompt,  # type: ignore[typeddict-item]
+                model=model,
+                with_diarization=with_diarization,  # type: ignore[typeddict-item]
+                num_speakers=num_speakers,  # type: ignore[typeddict-item]
+            ),
+            callback=callback,
+            request_options=request_options,
+        )
+        return AsyncSpeechToTextTranslateJob(job_id=response.job_id, client=self)
+
+    async def get_job(self, job_id: str) -> "AsyncSpeechToTextTranslateJob":
+        """
+        Get an existing Speech-to-Text-Translate job handle by job ID.
+
+        Parameters
+        ----------
+        job_id : str
+            The job ID of the previously created speech-to-text-translate job.
+
+        Returns
+        -------
+        AsyncSpeechToTextTranslateJob
+            A job handle which can be used to check status or retrieve results.
+        """
+        return AsyncSpeechToTextTranslateJob(job_id=job_id, client=self)