sarvamai 0.1.11a0__py3-none-any.whl → 0.1.11a2__py3-none-any.whl

sarvamai/__init__.py

@@ -3,7 +3,6 @@
 # isort: skip_file
 
 from .types import (
-    AudioCodec,
     AudioData,
     AudioMessage,
     AudioOutput,
@@ -25,6 +24,7 @@ from .types import (
     ConfigureConnection,
     ConfigureConnectionData,
     ConfigureConnectionDataOutputAudioBitrate,
+    ConfigureConnectionDataOutputAudioCodec,
     ConfigureConnectionDataSpeaker,
     ConfigureConnectionDataTargetLanguageCode,
     CreateChatCompletionResponse,
@@ -43,7 +43,6 @@ from .types import (
     FilesUploadResponse,
     FinishReason,
     FlushSignal,
-    Format,
     JobState,
     JobStatusV1,
     LanguageIdentificationResponse,
@@ -77,6 +76,7 @@ from .types import (
     TaskState,
     TextToSpeechLanguage,
     TextToSpeechModel,
+    TextToSpeechOutputAudioCodec,
     TextToSpeechResponse,
     TextToSpeechSpeaker,
     TimestampsModel,
@@ -185,7 +185,6 @@ from .version import __version__
 
 __all__ = [
     "AsyncSarvamAI",
-    "AudioCodec",
     "AudioData",
     "AudioDataParams",
     "AudioMessage",
@@ -226,6 +225,7 @@ __all__ = [
     "ConfigureConnection",
     "ConfigureConnectionData",
     "ConfigureConnectionDataOutputAudioBitrate",
+    "ConfigureConnectionDataOutputAudioCodec",
     "ConfigureConnectionDataParams",
     "ConfigureConnectionDataSpeaker",
     "ConfigureConnectionDataTargetLanguageCode",
@@ -261,7 +261,6 @@ __all__ = [
     "FlushSignal",
     "FlushSignalParams",
     "ForbiddenError",
-    "Format",
     "InternalServerError",
     "JobState",
     "JobStatusV1",
@@ -323,6 +322,7 @@ __all__ = [
     "TaskState",
     "TextToSpeechLanguage",
     "TextToSpeechModel",
+    "TextToSpeechOutputAudioCodec",
     "TextToSpeechResponse",
     "TextToSpeechResponseParams",
     "TextToSpeechSpeaker",

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/0.1.11a0",
+            "User-Agent": "sarvamai/0.1.11a2",
             "X-Fern-Language": "Python",
             "X-Fern-SDK-Name": "sarvamai",
-            "X-Fern-SDK-Version": "0.1.11a0",
+            "X-Fern-SDK-Version": "0.1.11a2",
             **(self.get_custom_headers() or {}),
         }
         headers["api-subscription-key"] = self.api_subscription_key

sarvamai/requests/configure_connection_data.py

@@ -1,9 +1,8 @@
 # This file was auto-generated by Fern from our API Definition.
 
-import typing
-
 import typing_extensions
 from ..types.configure_connection_data_output_audio_bitrate import ConfigureConnectionDataOutputAudioBitrate
+from ..types.configure_connection_data_output_audio_codec import ConfigureConnectionDataOutputAudioCodec
 from ..types.configure_connection_data_speaker import ConfigureConnectionDataSpeaker
 from ..types.configure_connection_data_target_language_code import ConfigureConnectionDataTargetLanguageCode
 
@@ -62,7 +61,7 @@ class ConfigureConnectionDataParams(typing_extensions.TypedDict):
     of mixed-language text. Default is false.
     """
 
-    output_audio_codec: typing_extensions.NotRequired[typing.Literal["mp3"]]
+    output_audio_codec: typing_extensions.NotRequired[ConfigureConnectionDataOutputAudioCodec]
     """
     Audio codec (currently supports MP3 only, optimized for real-time playback)
     """

sarvamai/speech_to_text_job/client.py

@@ -11,6 +11,7 @@ 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 AsyncRawSpeechToTextJobClient, RawSpeechToTextJobClient
+from .job import AsyncSpeechToTextJob, SpeechToTextJob
 
 # this is used as the default value for optional parameters
 OMIT = typing.cast(typing.Any, ...)
@@ -69,11 +70,15 @@ class SpeechToTextJobClient:
         )
         """
         _response = self._raw_client.initialise(
-            job_parameters=job_parameters, callback=callback, request_options=request_options
+            job_parameters=job_parameters,
+            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 bulk job V1
 
@@ -140,11 +145,17 @@ class SpeechToTextJobClient:
             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(
-        self, *, job_id: str, files: typing.Sequence[str], request_options: typing.Optional[RequestOptions] = None
+        self,
+        *,
+        job_id: str,
+        files: typing.Sequence[str],
+        request_options: typing.Optional[RequestOptions] = None,
     ) -> FilesUploadResponse:
         """
         Start a speech to text bulk job V1
@@ -175,11 +186,17 @@ class SpeechToTextJobClient:
             files=["files"],
         )
         """
-        _response = self._raw_client.get_upload_links(job_id=job_id, files=files, request_options=request_options)
+        _response = self._raw_client.get_upload_links(
+            job_id=job_id, files=files, request_options=request_options
+        )
         return _response.data
 
     def get_download_links(
-        self, *, job_id: str, files: typing.Sequence[str], request_options: typing.Optional[RequestOptions] = None
+        self,
+        *,
+        job_id: str,
+        files: typing.Sequence[str],
+        request_options: typing.Optional[RequestOptions] = None,
     ) -> FilesDownloadResponse:
         """
         Start a speech to text bulk job V1
@@ -210,9 +227,30 @@ class SpeechToTextJobClient:
             files=["files"],
         )
         """
-        _response = self._raw_client.get_download_links(job_id=job_id, files=files, request_options=request_options)
+        _response = self._raw_client.get_download_links(
+            job_id=job_id, files=files, request_options=request_options
+        )
         return _response.data
 
+    def create_job(
+        self,
+        job_parameters: SpeechToTextJobParametersParams,
+        callback: typing.Optional[BulkJobCallbackParams] = OMIT,
+        request_options: typing.Optional[RequestOptions] = None,
+    ) -> SpeechToTextJob:
+        response = self.initialise(
+            job_parameters=job_parameters,
+            callback=callback,
+            request_options=request_options,
+        )
+        return SpeechToTextJob(job_id=response.job_id, client=self)
+
+    def get_job(self, job_id: str) -> SpeechToTextJob:
+        """
+        Return a job handle for an existing Speech-to-Text job.
+        """
+        return SpeechToTextJob(job_id=job_id, client=self)
+
 
 class AsyncSpeechToTextJobClient:
     def __init__(self, *, client_wrapper: AsyncClientWrapper):
@@ -275,11 +313,15 @@ class AsyncSpeechToTextJobClient:
         asyncio.run(main())
         """
         _response = await self._raw_client.initialise(
-            job_parameters=job_parameters, callback=callback, request_options=request_options
+            job_parameters=job_parameters,
+            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 bulk job V1
 
@@ -315,7 +357,9 @@ class AsyncSpeechToTextJobClient:
 
         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(
@@ -362,11 +406,17 @@ class AsyncSpeechToTextJobClient:
 
         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(
-        self, *, job_id: str, files: typing.Sequence[str], request_options: typing.Optional[RequestOptions] = None
+        self,
+        *,
+        job_id: str,
+        files: typing.Sequence[str],
+        request_options: typing.Optional[RequestOptions] = None,
     ) -> FilesUploadResponse:
         """
         Start a speech to text bulk job V1
@@ -405,11 +455,17 @@ class AsyncSpeechToTextJobClient:
 
         asyncio.run(main())
         """
-        _response = await self._raw_client.get_upload_links(job_id=job_id, files=files, request_options=request_options)
+        _response = await self._raw_client.get_upload_links(
+            job_id=job_id, files=files, request_options=request_options
+        )
         return _response.data
 
     async def get_download_links(
-        self, *, job_id: str, files: typing.Sequence[str], request_options: typing.Optional[RequestOptions] = None
+        self,
+        *,
+        job_id: str,
+        files: typing.Sequence[str],
+        request_options: typing.Optional[RequestOptions] = None,
     ) -> FilesDownloadResponse:
         """
         Start a speech to text bulk job V1
@@ -452,3 +508,22 @@ class AsyncSpeechToTextJobClient:
             job_id=job_id, files=files, request_options=request_options
         )
         return _response.data
+
+    async def create_job(
+        self,
+        job_parameters: SpeechToTextJobParametersParams,
+        callback: typing.Optional[BulkJobCallbackParams] = OMIT,
+        request_options: typing.Optional[RequestOptions] = None,
+    ) -> "AsyncSpeechToTextJob":
+        response = await self.initialise(
+            job_parameters=job_parameters,
+            callback=callback,
+            request_options=request_options,
+        )
+        return AsyncSpeechToTextJob(job_id=response.job_id, client=self)
+
+    async def get_job(self, job_id: str) -> "AsyncSpeechToTextJob":
+        """
+        Return a job handle for an existing Speech-to-Text job.
+        """
+        return AsyncSpeechToTextJob(job_id=job_id, client=self)

sarvamai/speech_to_text_job/job.py

@@ -0,0 +1,468 @@
+import asyncio
+import os
+import time
+import typing
+import httpx
+
+from ..types import JobStatusV1
+
+if typing.TYPE_CHECKING:
+    from .client import AsyncSpeechToTextJobClient, SpeechToTextJobClient
+
+
+class AsyncSpeechToTextJob:
+    def __init__(self, job_id: str, client: "AsyncSpeechToTextJobClient"):
+        """
+        Initialize the asynchronous speech-to-text job.
+
+        Parameters
+        ----------
+        job_id : str
+            The unique job identifier returned from a previous job initialization.
+
+        client : AsyncSpeechToTextJobClient
+            The async client instance used to create the job.
+
+            !!! important
+                This must be the **same client instance** that was used to initialize
+                the job originally, as it contains the subscription key and configuration
+                required to authenticate and manage the job.
+
+        """
+        self._job_id = job_id
+        self._client = client
+
+    @property
+    def job_id(self) -> str:
+        """
+        Returns the job ID associated with this job instance.
+
+        Returns
+        -------
+        str
+        """
+        return self._job_id
+
+    async def upload_files(self, file_paths: typing.Sequence[str]) -> bool:
+        """
+        Upload input audio files for the speech-to-text job.
+
+        Parameters
+        ----------
+        file_paths : Sequence[str]
+            List of full paths to local audio files.
+
+        Returns
+        -------
+        bool
+            True if all files are uploaded successfully.
+        """
+        upload_links = await self._client.get_upload_links(
+            job_id=self._job_id,
+            files=[os.path.basename(p) for p in file_paths],
+        )
+        async with httpx.AsyncClient() as session:
+            for path in file_paths:
+                file_name = os.path.basename(path)
+                url = upload_links.upload_urls[file_name].file_url
+                with open(path, "rb") as f:
+                    response = await session.put(
+                        url,
+                        content=f.read(),
+                        headers={
+                            "x-ms-blob-type": "BlockBlob",
+                            "Content-Type": "audio/wav",
+                        },
+                    )
+                if response.status_code != 201:
+                    raise RuntimeError(
+                        f"Upload failed for {file_name}: {response.status_code}"
+                    )
+        return True
+
+    async def wait_until_complete(
+        self, poll_interval: int = 5, timeout: int = 600
+    ) -> JobStatusV1:
+        """
+        Polls job status until it completes or fails.
+
+        Parameters
+        ----------
+        poll_interval : int, optional
+            Time in seconds between polling attempts (default is 5).
+
+        timeout : int, optional
+            Maximum time to wait for completion in seconds (default is 600).
+
+        Returns
+        -------
+        JobStatusV1
+            Final job status.
+
+        Raises
+        ------
+        TimeoutError
+            If the job does not complete within the given timeout.
+        """
+        start = asyncio.get_event_loop().time()
+        while True:
+            status = await self.get_status()
+            state = status.job_state.lower()
+            if state in {"completed", "failed"}:
+                return status
+            if asyncio.get_event_loop().time() - start > timeout:
+                raise TimeoutError(
+                    f"Job {self._job_id} did not complete within {timeout} seconds."
+                )
+            await asyncio.sleep(poll_interval)
+
+    async def get_output_mappings(self) -> typing.List[typing.Dict[str, str]]:
+        """
+        Get the mapping of input files to their corresponding output files.
+
+        Returns
+        -------
+        List[Dict[str, str]]
+            List of mappings with keys 'input_file' and 'output_file'.
+        """
+        job_status = await self.get_status()
+        return [
+            {
+                "input_file": detail.inputs[0].file_name,
+                "output_file": detail.outputs[0].file_name,
+            }
+            for detail in (job_status.job_details or [])
+            if detail.inputs and detail.outputs
+        ]
+
+    async def download_outputs(self, output_dir: str) -> bool:
+        """
+        Download output files to the specified directory.
+
+        Parameters
+        ----------
+        output_dir : str
+            Local directory where outputs will be saved.
+
+        Returns
+        -------
+        bool
+            True if all files downloaded successfully.
+
+        Raises
+        ------
+        RuntimeError
+            If a file fails to download.
+        """
+        mappings = await self.get_output_mappings()
+        file_names = [m["output_file"] for m in mappings]
+        download_links = await self._client.get_download_links(
+            job_id=self._job_id, files=file_names
+        )
+
+        os.makedirs(output_dir, exist_ok=True)
+        async with httpx.AsyncClient() as session:
+            for m in mappings:
+                url = download_links.download_urls[m["output_file"]].file_url
+                response = await session.get(url)
+                if response.status_code != 200:
+                    raise RuntimeError(
+                        f"Download failed for {m['output_file']}: {response.status_code}"
+                    )
+                output_path = os.path.join(output_dir, f"{m['input_file']}.json")
+                with open(output_path, "wb") as f:
+                    f.write(response.content)
+        return True
+
+    async def get_status(self) -> JobStatusV1:
+        """
+        Retrieve the current status of the job.
+
+        Returns
+        -------
+        JobStatusV1
+        """
+        return await self._client.get_status(self._job_id)
+
+    async def start(self) -> JobStatusV1:
+        """
+        Start the speech-to-text job processing.
+
+        Returns
+        -------
+        JobStatusV1
+        """
+        return await self._client.start(job_id=self._job_id)
+
+    async def exists(self) -> bool:
+        """
+        Check if the job exists in the system.
+
+        Returns
+        -------
+        bool
+        """
+        try:
+            await self.get_status()
+            return True
+        except httpx.HTTPStatusError:
+            return False
+
+    async def is_complete(self) -> bool:
+        """
+        Check if the job is either completed or failed.
+
+        Returns
+        -------
+        bool
+        """
+        state = (await self.get_status()).job_state.lower()
+        return state in {"completed", "failed"}
+
+    async def is_successful(self) -> bool:
+        """
+        Check if the job completed successfully.
+
+        Returns
+        -------
+        bool
+        """
+        return (await self.get_status()).job_state.lower() == "completed"
+
+    async def is_failed(self) -> bool:
+        """
+        Check if the job has failed.
+
+        Returns
+        -------
+        bool
+        """
+        return (await self.get_status()).job_state.lower() == "failed"
+
+
+class SpeechToTextJob:
+    def __init__(self, job_id: str, client: "SpeechToTextJobClient"):
+        """
+        Initialize the synchronous speech-to-text job.
+
+        Parameters
+        ----------
+        job_id : str
+            The unique job identifier returned from a previous job initialization.
+
+        client : SpeechToTextJobClient
+            The client instance used to create the job.
+
+            !!! important
+                This must be the **same client instance** that was used to initialize
+                the job originally, as it contains the subscription key and configuration
+                required to authenticate and manage the job.
+
+        """
+        self._job_id = job_id
+        self._client = client
+
+    @property
+    def job_id(self) -> str:
+        """
+        Returns the job ID associated with this job instance.
+
+        Returns
+        -------
+        str
+        """
+        return self._job_id
+
+    def upload_files(self, file_paths: typing.Sequence[str]) -> bool:
+        """
+        Upload input audio files for the speech-to-text job.
+
+        Parameters
+        ----------
+        file_paths : Sequence[str]
+            List of full paths to local audio files.
+
+        Returns
+        -------
+        bool
+            True if all files are uploaded successfully.
+        """
+        upload_links = self._client.get_upload_links(
+            job_id=self._job_id, files=[os.path.basename(p) for p in file_paths]
+        )
+        with httpx.Client() as client:
+            for path in file_paths:
+                file_name = os.path.basename(path)
+                url = upload_links.upload_urls[file_name].file_url
+                with open(path, "rb") as f:
+                    response = client.put(
+                        url,
+                        content=f,
+                        headers={
+                            "x-ms-blob-type": "BlockBlob",
+                            "Content-Type": "audio/wav",
+                        },
+                    )
+                if response.status_code != 201:
+                    raise RuntimeError(
+                        f"Upload failed for {file_name}: {response.status_code}"
+                    )
+        return True
+
+    def wait_until_complete(
+        self, poll_interval: int = 5, timeout: int = 600
+    ) -> JobStatusV1:
+        """
+        Polls job status until it completes or fails.
+
+        Parameters
+        ----------
+        poll_interval : int, optional
+            Time in seconds between polling attempts (default is 5).
+
+        timeout : int, optional
+            Maximum time to wait for completion in seconds (default is 600).
+
+        Returns
+        -------
+        JobStatusV1
+            Final job status.
+
+        Raises
+        ------
+        TimeoutError
+            If the job does not complete within the given timeout.
+        """
+        start = time.monotonic()
+        while True:
+            status = self.get_status()
+            state = status.job_state.lower()
+            if state in {"completed", "failed"}:
+                return status
+            if time.monotonic() - start > timeout:
+                raise TimeoutError(
+                    f"Job {self._job_id} did not complete within {timeout} seconds."
+                )
+            time.sleep(poll_interval)
+
+    def get_output_mappings(self) -> typing.List[typing.Dict[str, str]]:
+        """
+        Get the mapping of input files to their corresponding output files.
+
+        Returns
+        -------
+        List[Dict[str, str]]
+            List of mappings with keys 'input_file' and 'output_file'.
+        """
+        job_status = self.get_status()
+        return [
+            {
+                "input_file": detail.inputs[0].file_name,
+                "output_file": detail.outputs[0].file_name,
+            }
+            for detail in (job_status.job_details or [])
+            if detail.inputs and detail.outputs
+        ]
+
+    def download_outputs(self, output_dir: str) -> bool:
+        """
+        Download output files to the specified directory.
+
+        Parameters
+        ----------
+        output_dir : str
+            Local directory where outputs will be saved.
+
+        Returns
+        -------
+        bool
+            True if all files downloaded successfully.
+
+        Raises
+        ------
+        RuntimeError
+            If a file fails to download.
+        """
+        mappings = self.get_output_mappings()
+        file_names = [m["output_file"] for m in mappings]
+        download_links = self._client.get_download_links(
+            job_id=self._job_id, files=file_names
+        )
+
+        os.makedirs(output_dir, exist_ok=True)
+        with httpx.Client() as client:
+            for m in mappings:
+                url = download_links.download_urls[m["output_file"]].file_url
+                response = client.get(url)
+                if response.status_code != 200:
+                    raise RuntimeError(
+                        f"Download failed for {m['output_file']}: {response.status_code}"
+                    )
+                output_path = os.path.join(output_dir, f"{m['input_file']}.json")
+                with open(output_path, "wb") as f:
+                    f.write(response.content)
+        return True
+
+    def get_status(self) -> JobStatusV1:
+        """
+        Retrieve the current status of the job.
+
+        Returns
+        -------
+        JobStatusV1
+        """
+        return self._client.get_status(self._job_id)
+
+    def start(self) -> JobStatusV1:
+        """
+        Start the speech-to-text job processing.
+
+        Returns
+        -------
+        JobStatusV1
+        """
+        return self._client.start(job_id=self._job_id)
+
+    def exists(self) -> bool:
+        """
+        Check if the job exists in the system.
+
+        Returns
+        -------
+        bool
+        """
+        try:
+            self.get_status()
+            return True
+        except httpx.HTTPStatusError:
+            return False
+
+    def is_complete(self) -> bool:
+        """
+        Check if the job is either completed or failed.
+
+        Returns
+        -------
+        bool
+        """
+        return self.get_status().job_state.lower() in {"completed", "failed"}
+
+    def is_successful(self) -> bool:
+        """
+        Check if the job completed successfully.
+
+        Returns
+        -------
+        bool
+        """
+        return self.get_status().job_state.lower() == "completed"
+
+    def is_failed(self) -> bool:
+        """
+        Check if the job has failed.
+
+        Returns
+        -------
+        bool
+        """
+        return self.get_status().job_state.lower() == "failed"

sarvamai/text_to_speech/client.py

@@ -4,10 +4,10 @@ import typing
 
 from ..core.client_wrapper import AsyncClientWrapper, SyncClientWrapper
 from ..core.request_options import RequestOptions
-from ..types.audio_codec import AudioCodec
 from ..types.speech_sample_rate import SpeechSampleRate
 from ..types.text_to_speech_language import TextToSpeechLanguage
 from ..types.text_to_speech_model import TextToSpeechModel
+from ..types.text_to_speech_output_audio_codec import TextToSpeechOutputAudioCodec
 from ..types.text_to_speech_response import TextToSpeechResponse
 from ..types.text_to_speech_speaker import TextToSpeechSpeaker
 from .raw_client import AsyncRawTextToSpeechClient, RawTextToSpeechClient
@@ -43,7 +43,7 @@ class TextToSpeechClient:
         speech_sample_rate: typing.Optional[SpeechSampleRate] = OMIT,
         enable_preprocessing: typing.Optional[bool] = OMIT,
         model: typing.Optional[TextToSpeechModel] = OMIT,
-        audio_codec: typing.Optional[AudioCodec] = OMIT,
+        output_audio_codec: typing.Optional[TextToSpeechOutputAudioCodec] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> TextToSpeechResponse:
         """
@@ -87,7 +87,7 @@ class TextToSpeechClient:
         model : typing.Optional[TextToSpeechModel]
             Specifies the model to use for text-to-speech conversion. Default is bulbul:v2.
 
-        audio_codec : typing.Optional[AudioCodec]
+        output_audio_codec : typing.Optional[TextToSpeechOutputAudioCodec]
             Specifies the audio codec for the output audio file. Different codecs offer various compression and quality characteristics.
 
         request_options : typing.Optional[RequestOptions]
@@ -120,7 +120,7 @@ class TextToSpeechClient:
             speech_sample_rate=speech_sample_rate,
             enable_preprocessing=enable_preprocessing,
             model=model,
-            audio_codec=audio_codec,
+            output_audio_codec=output_audio_codec,
             request_options=request_options,
         )
         return _response.data
@@ -153,7 +153,7 @@ class AsyncTextToSpeechClient:
         speech_sample_rate: typing.Optional[SpeechSampleRate] = OMIT,
         enable_preprocessing: typing.Optional[bool] = OMIT,
         model: typing.Optional[TextToSpeechModel] = OMIT,
-        audio_codec: typing.Optional[AudioCodec] = OMIT,
+        output_audio_codec: typing.Optional[TextToSpeechOutputAudioCodec] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> TextToSpeechResponse:
         """
@@ -197,7 +197,7 @@ class AsyncTextToSpeechClient:
         model : typing.Optional[TextToSpeechModel]
             Specifies the model to use for text-to-speech conversion. Default is bulbul:v2.
 
-        audio_codec : typing.Optional[AudioCodec]
+        output_audio_codec : typing.Optional[TextToSpeechOutputAudioCodec]
             Specifies the audio codec for the output audio file. Different codecs offer various compression and quality characteristics.
 
         request_options : typing.Optional[RequestOptions]
@@ -238,7 +238,7 @@ class AsyncTextToSpeechClient:
             speech_sample_rate=speech_sample_rate,
             enable_preprocessing=enable_preprocessing,
             model=model,
-            audio_codec=audio_codec,
+            output_audio_codec=output_audio_codec,
             request_options=request_options,
         )
         return _response.data

sarvamai/text_to_speech/raw_client.py

@@ -13,10 +13,10 @@ from ..errors.forbidden_error import ForbiddenError
 from ..errors.internal_server_error import InternalServerError
 from ..errors.too_many_requests_error import TooManyRequestsError
 from ..errors.unprocessable_entity_error import UnprocessableEntityError
-from ..types.audio_codec import AudioCodec
 from ..types.speech_sample_rate import SpeechSampleRate
 from ..types.text_to_speech_language import TextToSpeechLanguage
 from ..types.text_to_speech_model import TextToSpeechModel
+from ..types.text_to_speech_output_audio_codec import TextToSpeechOutputAudioCodec
 from ..types.text_to_speech_response import TextToSpeechResponse
 from ..types.text_to_speech_speaker import TextToSpeechSpeaker
 
@@ -40,7 +40,7 @@ class RawTextToSpeechClient:
         speech_sample_rate: typing.Optional[SpeechSampleRate] = OMIT,
         enable_preprocessing: typing.Optional[bool] = OMIT,
         model: typing.Optional[TextToSpeechModel] = OMIT,
-        audio_codec: typing.Optional[AudioCodec] = OMIT,
+        output_audio_codec: typing.Optional[TextToSpeechOutputAudioCodec] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> HttpResponse[TextToSpeechResponse]:
         """
@@ -84,7 +84,7 @@ class RawTextToSpeechClient:
         model : typing.Optional[TextToSpeechModel]
             Specifies the model to use for text-to-speech conversion. Default is bulbul:v2.
 
-        audio_codec : typing.Optional[AudioCodec]
+        output_audio_codec : typing.Optional[TextToSpeechOutputAudioCodec]
             Specifies the audio codec for the output audio file. Different codecs offer various compression and quality characteristics.
 
         request_options : typing.Optional[RequestOptions]
@@ -109,7 +109,7 @@ class RawTextToSpeechClient:
                 "speech_sample_rate": speech_sample_rate,
                 "enable_preprocessing": enable_preprocessing,
                 "model": model,
-                "audio_codec": audio_codec,
+                "output_audio_codec": output_audio_codec,
             },
             headers={
                 "content-type": "application/json",
@@ -204,7 +204,7 @@ class AsyncRawTextToSpeechClient:
         speech_sample_rate: typing.Optional[SpeechSampleRate] = OMIT,
         enable_preprocessing: typing.Optional[bool] = OMIT,
         model: typing.Optional[TextToSpeechModel] = OMIT,
-        audio_codec: typing.Optional[AudioCodec] = OMIT,
+        output_audio_codec: typing.Optional[TextToSpeechOutputAudioCodec] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> AsyncHttpResponse[TextToSpeechResponse]:
         """
@@ -248,7 +248,7 @@ class AsyncRawTextToSpeechClient:
         model : typing.Optional[TextToSpeechModel]
             Specifies the model to use for text-to-speech conversion. Default is bulbul:v2.
 
-        audio_codec : typing.Optional[AudioCodec]
+        output_audio_codec : typing.Optional[TextToSpeechOutputAudioCodec]
             Specifies the audio codec for the output audio file. Different codecs offer various compression and quality characteristics.
 
         request_options : typing.Optional[RequestOptions]
@@ -273,7 +273,7 @@ class AsyncRawTextToSpeechClient:
                 "speech_sample_rate": speech_sample_rate,
                 "enable_preprocessing": enable_preprocessing,
                 "model": model,
-                "audio_codec": audio_codec,
+                "output_audio_codec": output_audio_codec,
             },
             headers={
                 "content-type": "application/json",

sarvamai/types/__init__.py

@@ -2,7 +2,6 @@
 
 # isort: skip_file
 
-from .audio_codec import AudioCodec
 from .audio_data import AudioData
 from .audio_message import AudioMessage
 from .audio_output import AudioOutput
@@ -26,6 +25,7 @@ from .config_message import ConfigMessage
 from .configure_connection import ConfigureConnection
 from .configure_connection_data import ConfigureConnectionData
 from .configure_connection_data_output_audio_bitrate import ConfigureConnectionDataOutputAudioBitrate
+from .configure_connection_data_output_audio_codec import ConfigureConnectionDataOutputAudioCodec
 from .configure_connection_data_speaker import ConfigureConnectionDataSpeaker
 from .configure_connection_data_target_language_code import ConfigureConnectionDataTargetLanguageCode
 from .create_chat_completion_response import CreateChatCompletionResponse
@@ -44,7 +44,6 @@ from .files_request import FilesRequest
 from .files_upload_response import FilesUploadResponse
 from .finish_reason import FinishReason
 from .flush_signal import FlushSignal
-from .format import Format
 from .job_state import JobState
 from .job_status_v_1 import JobStatusV1
 from .language_identification_response import LanguageIdentificationResponse
@@ -78,6 +77,7 @@ from .task_file_details import TaskFileDetails
 from .task_state import TaskState
 from .text_to_speech_language import TextToSpeechLanguage
 from .text_to_speech_model import TextToSpeechModel
+from .text_to_speech_output_audio_codec import TextToSpeechOutputAudioCodec
 from .text_to_speech_response import TextToSpeechResponse
 from .text_to_speech_speaker import TextToSpeechSpeaker
 from .timestamps_model import TimestampsModel
@@ -94,7 +94,6 @@ from .transliterate_source_language import TransliterateSourceLanguage
 from .transliteration_response import TransliterationResponse
 
 __all__ = [
-    "AudioCodec",
     "AudioData",
     "AudioMessage",
     "AudioOutput",
@@ -116,6 +115,7 @@ __all__ = [
     "ConfigureConnection",
     "ConfigureConnectionData",
     "ConfigureConnectionDataOutputAudioBitrate",
+    "ConfigureConnectionDataOutputAudioCodec",
     "ConfigureConnectionDataSpeaker",
     "ConfigureConnectionDataTargetLanguageCode",
     "CreateChatCompletionResponse",
@@ -134,7 +134,6 @@ __all__ = [
     "FilesUploadResponse",
     "FinishReason",
     "FlushSignal",
-    "Format",
     "JobState",
     "JobStatusV1",
     "LanguageIdentificationResponse",
@@ -168,6 +167,7 @@ __all__ = [
     "TaskState",
     "TextToSpeechLanguage",
     "TextToSpeechModel",
+    "TextToSpeechOutputAudioCodec",
     "TextToSpeechResponse",
     "TextToSpeechSpeaker",
     "TimestampsModel",

sarvamai/types/configure_connection_data.py

@@ -5,6 +5,7 @@ import typing
 import pydantic
 from ..core.pydantic_utilities import IS_PYDANTIC_V2, UniversalBaseModel
 from .configure_connection_data_output_audio_bitrate import ConfigureConnectionDataOutputAudioBitrate
+from .configure_connection_data_output_audio_codec import ConfigureConnectionDataOutputAudioCodec
 from .configure_connection_data_speaker import ConfigureConnectionDataSpeaker
 from .configure_connection_data_target_language_code import ConfigureConnectionDataTargetLanguageCode
 
@@ -63,7 +64,7 @@ class ConfigureConnectionData(UniversalBaseModel):
     of mixed-language text. Default is false.
     """
 
-    output_audio_codec: typing.Optional[typing.Literal["mp3"]] = pydantic.Field(default=None)
+    output_audio_codec: typing.Optional[ConfigureConnectionDataOutputAudioCodec] = pydantic.Field(default=None)
     """
     Audio codec (currently supports MP3 only, optimized for real-time playback)
     """

sarvamai/types/configure_connection_data_output_audio_codec.py

@@ -0,0 +1,7 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+ConfigureConnectionDataOutputAudioCodec = typing.Union[
+    typing.Literal["linear16", "mulaw", "alaw", "opus", "flac", "aac", "wav"], typing.Any
+]

sarvamai/types/text_to_speech_output_audio_codec.py

@@ -0,0 +1,7 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+TextToSpeechOutputAudioCodec = typing.Union[
+    typing.Literal["mp3", "linear16", "mulaw", "alaw", "opus", "flac", "aac", "wav"], typing.Any
+]

{sarvamai-0.1.11a0.dist-info → sarvamai-0.1.11a2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sarvamai
-Version: 0.1.11a0
+Version: 0.1.11a2
 Summary: 
 Requires-Python: >=3.8,<4.0
 Classifier: Intended Audience :: Developers

{sarvamai-0.1.11a0.dist-info → sarvamai-0.1.11a2.dist-info}/RECORD

@@ -1,11 +1,11 @@
-sarvamai/__init__.py,sha256=3GBj0sRY6_5r171hnXCnlBqmNWAwjvsuq9JTnNnmIRY,10216
+sarvamai/__init__.py,sha256=StGifG-VL1O3Xam5IR6bQa_8QfRA46pDQ0BgidTQx4U,10318
 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=5YC2fxVENOxQXoY-t3n8qZ0aQ9UasDjFRzBZw8ce9OQ,7861
 sarvamai/core/__init__.py,sha256=YE2CtXeASe1RAbaI39twKWYKCuT4tW5is9HWHhJjR_g,1653
 sarvamai/core/api_error.py,sha256=44vPoTyWN59gonCIZMdzw7M1uspygiLnr3GNFOoVL2Q,614
-sarvamai/core/client_wrapper.py,sha256=bU4r7rXjgAN2ysY93DFLuhVyXCrituXocTBpmrc4wCo,2570
+sarvamai/core/client_wrapper.py,sha256=iqRsEvtadYxKdwKY3YUlCec4a0SppCuIqnrLItNLO2c,2570
 sarvamai/core/datetime_utils.py,sha256=nBys2IsYrhPdszxGKCNRPSOCwa-5DWOHG95FB8G9PKo,1047
 sarvamai/core/events.py,sha256=j7VWXgMpOsjCXdzY22wIhI7Q-v5InZ4WchRzA88x_Sk,856
 sarvamai/core/file.py,sha256=d4NNbX8XvXP32z8KpK2Xovv33nFfruIrpz0QWxlgpZk,2663
@@ -45,7 +45,7 @@ sarvamai/requests/choice.py,sha256=uulX4MZUoThEMcD3a80o_3V5YpnpqN8DfPaNZWVz-1o,8
 sarvamai/requests/completion_usage.py,sha256=LbZV-RxcxKdCAYqhCiaRtSFF3VwMJq71A989Z1rm-I8,428
 sarvamai/requests/config_message.py,sha256=EpYioGvDhCXDMvGH7Q1F7448zJzoHmlkQ1owoNGbWAw,383
 sarvamai/requests/configure_connection.py,sha256=a-foQtLxArL4CulvKEdeebbRqmS1GRmko3MZdnHVPEk,716
-sarvamai/requests/configure_connection_data.py,sha256=Niil2OrVBzQEtmWFn1JC-StLVp6WzzRIsu2i_M_8_44,2908
+sarvamai/requests/configure_connection_data.py,sha256=lRk_4rYPQLLlwS2HXjQ9Abxdf98_DuOOja-VkrIR44Q,3016
 sarvamai/requests/create_chat_completion_response.py,sha256=TqS9u5_WVWMok_NreT4TeOsLJQeybPkbJm45Q0Zxw30,857
 sarvamai/requests/diarized_entry.py,sha256=gbXB4D_r5_Q8gs1arRKjxPeFcYg16dVDLcg2VhxmKQA,462
 sarvamai/requests/diarized_transcript.py,sha256=X-znuJ45oqwXzVyJumBHSqVGLz6JnoYFZmluQlEpEAw,323
@@ -86,7 +86,8 @@ sarvamai/speech_to_text/__init__.py,sha256=_VhToAyIt_5axN6CLJwtxg3-CO7THa_23pbUz
 sarvamai/speech_to_text/client.py,sha256=lp2G2fI9SUbeOBBE1S5tjcp-Xb8wIhAuVadLKwXveh8,11003
 sarvamai/speech_to_text/raw_client.py,sha256=A_56vEVeJdyttVJRiFxTMJ4n-s4l_PS8rI1DiLZlOmc,25331
 sarvamai/speech_to_text_job/__init__.py,sha256=_VhToAyIt_5axN6CLJwtxg3-CO7THa_23pbUzqhXJa4,85
-sarvamai/speech_to_text_job/client.py,sha256=kmYoK39CK9TlLwSRU3sLzTbP4QzqmHbk3ehC8oUASmA,12468
+sarvamai/speech_to_text_job/client.py,sha256=DJhCmYQ1EYbmrhLalsqEv_3rZYJWTsT-W2wQrU6IaXU,14275
+sarvamai/speech_to_text_job/job.py,sha256=6h9hrpsKX21ql-umxEThtx8gU38cZj25WuBY9iBdnbI,13744
 sarvamai/speech_to_text_job/raw_client.py,sha256=v14drcQLAmpqozRUNKmw1F9j3omieMPC8R88Th1BID8,48055
 sarvamai/speech_to_text_streaming/__init__.py,sha256=q7QygMmZCHJ-4FMhhL_6XNV_dsqlIFRCO1iSxoyxaaY,437
 sarvamai/speech_to_text_streaming/client.py,sha256=WdkzZxKMdnQ2hHv9hzJlfSNggRJLKFljRiC7695Jcog,8224
@@ -109,14 +110,13 @@ sarvamai/text/__init__.py,sha256=_VhToAyIt_5axN6CLJwtxg3-CO7THa_23pbUzqhXJa4,85
 sarvamai/text/client.py,sha256=2kA0Gxfi-r52zMQdqRRD811014alzlHB_FANkp3Kn_c,30595
 sarvamai/text/raw_client.py,sha256=7xYmJA50kTKy_gj8tkAPckKp2djHB37zOdm0_icbMb8,48695
 sarvamai/text_to_speech/__init__.py,sha256=_VhToAyIt_5axN6CLJwtxg3-CO7THa_23pbUzqhXJa4,85
-sarvamai/text_to_speech/client.py,sha256=ERj6e9ZBkdv16D344Z3C2EeqBbGezZx6GuzANoXHKkY,9449
-sarvamai/text_to_speech/raw_client.py,sha256=WCbuL3i7FRL-vwCRER9Y47wPGRMzSzoeQHHeV2E7l78,15401
+sarvamai/text_to_speech/client.py,sha256=iwrQNfoMgCSOgvztTIXtLHQmSmn0RInwt5RSo9TwdtA,9617
+sarvamai/text_to_speech/raw_client.py,sha256=si_aSjMR7SocIpKZFoVYqBmaIDuRm_6vxTM0dJ73PEo,15569
 sarvamai/text_to_speech_streaming/__init__.py,sha256=_VhToAyIt_5axN6CLJwtxg3-CO7THa_23pbUzqhXJa4,85
 sarvamai/text_to_speech_streaming/client.py,sha256=geTF5xy-batzO12XVt0sPw_XJCi7-m2sDFK_B7SL7qc,6088
 sarvamai/text_to_speech_streaming/raw_client.py,sha256=asOcNw1WAViOiXDVWH4sxWSXGVoLwAOh9vUtq_xralA,5269
 sarvamai/text_to_speech_streaming/socket_client.py,sha256=NEcijnvjuNcWfzqpBi-xWsXVkL0NPq6EGAkEjnaq9hw,13909
-sarvamai/types/__init__.py,sha256=Vdknm-xJ1MfSywqApXjWkjATf_ZPfWFm4mI2-2RSRBs,7389
-sarvamai/types/audio_codec.py,sha256=9qNJc1SdIP4IKHYwclqqtU9810qTaJtsSGx22AqKhKY,200
+sarvamai/types/__init__.py,sha256=gedZvTQHvAgl07VqBNMFY-su7Vnys_wBvCanZXD6Ews,7551
 sarvamai/types/audio_data.py,sha256=rgOukLkLNJ_HBBVE2g5dfEL2CWjRoGiMvCtpq0qTB1Y,829
 sarvamai/types/audio_message.py,sha256=sB4EgkWkWJzipYXobkmM9AYZTTZtCpg_ySKssUeznUE,560
 sarvamai/types/audio_output.py,sha256=Eq-YUZa1mSDwt7bax2c4Vv2gBlyM_JBJWzHhTAhFSko,621
@@ -133,8 +133,9 @@ sarvamai/types/choice.py,sha256=uXBCsjWP9VK3XWQWZUeI4EnU10w0G9nAfKn2tJZvxko,1244
 sarvamai/types/completion_usage.py,sha256=xYQGlQUbKqsksuV73H-1ajjfT5M7w47eLfdWXSlrI5M,843
 sarvamai/types/config_message.py,sha256=sGrT-qYTRqLVfIo5nRUuRlqPtPVmiAkUAnaMtlmQYCU,778
 sarvamai/types/configure_connection.py,sha256=SnSNk02gQqP8e4VB4y88jjeFQ4ClpImjGLn2ANI8cZ4,1058
-sarvamai/types/configure_connection_data.py,sha256=brMO-Z1TDq3oTJ22m1icBkkmnd9k67p_DzecnMcqNko,3421
+sarvamai/types/configure_connection_data.py,sha256=uXC7fhNJWCpaKc2Vrz2DNpUxx1gN3PwAoDL-H8L401A,3537
 sarvamai/types/configure_connection_data_output_audio_bitrate.py,sha256=h00YvKLxsZC8L3__rH4XH53nN_GY40UElW1EjysCwUs,208
+sarvamai/types/configure_connection_data_output_audio_codec.py,sha256=ddd-MjgmKE0e5-TPgPKclBu4h9WLC5g3kL8Ap_91i50,228
 sarvamai/types/configure_connection_data_speaker.py,sha256=SzyAiK5LynXwb9KniaO2qoOLY-II3-PMZbRuIsQ9shw,230
 sarvamai/types/configure_connection_data_target_language_code.py,sha256=jrU1EblAtDYbybUO1KUkHhevmlSBj2AQxX13ii3QhAQ,275
 sarvamai/types/create_chat_completion_response.py,sha256=4nEzeWzHGW1_BmRAtOuGsbRZ0ojNgnzJSMUFyYuYviw,1285
@@ -153,7 +154,6 @@ sarvamai/types/files_request.py,sha256=Jh8xPjoOTjY7DOE2EieoRqtkWkYxz9j-BP8TvWxuR
 sarvamai/types/files_upload_response.py,sha256=wRntZyh1-LGpo4-x_986Nv2A9rv9asDx93pqoQxNpTY,804
 sarvamai/types/finish_reason.py,sha256=PBWtBNkX4FMaODmlUehpF6qLB5uH_zR-Mw3M4uhIB6U,209
 sarvamai/types/flush_signal.py,sha256=N7MJWb658KoxRpFN9cIbyQGY45zZcg8YCou3E1v--9o,759
-sarvamai/types/format.py,sha256=57LicD0XLqW4D1QEnZWsWGifzRy1GV9P5utKPXLoxtg,144
 sarvamai/types/job_state.py,sha256=H6Zph2mIcjsd3upEDt1VzIEORkEpnIDs0kH8BvIyrow,189
 sarvamai/types/job_status_v_1.py,sha256=i1xopAptPVbGGIUcjKWgjAzSKwLVy6y4oGVEYcOA1P0,1798
 sarvamai/types/language_identification_response.py,sha256=jG4ZQ6KQHCiEDqC51OniOwiRdW14Fbz22bbTsUDp_kc,1483
@@ -187,6 +187,7 @@ sarvamai/types/task_file_details.py,sha256=oJV7zaUVrbqqw-If-2_V1aLk28qW0ZbeIDtIp
 sarvamai/types/task_state.py,sha256=fSrmD00Goi0J6s9hzqcFqz3Fkh37diBYpxnz4FkwHdU,182
 sarvamai/types/text_to_speech_language.py,sha256=T5-rP93WyJwkdJTmNZuvNkkoVtgU0G25a8R2F3OwRZ4,254
 sarvamai/types/text_to_speech_model.py,sha256=qRkpGCcfrLD45l499cBUcBgZDo_qKPZtFxA7wPbp1NQ,128
+sarvamai/types/text_to_speech_output_audio_codec.py,sha256=lW-w0RIQiLZGdVmUgg20yYEblvaZ9AvvqzE6o7xRhWQ,224
 sarvamai/types/text_to_speech_response.py,sha256=Yzvwvwm65IR2vUzxZws9OLBW0GgB6bbmvjClqSuZzdg,742
 sarvamai/types/text_to_speech_speaker.py,sha256=300mXxDKnWV9O7ccuO8EkLooAu2-y2ZdqDynrckaHew,219
 sarvamai/types/timestamps_model.py,sha256=ZlqcxYNtAcm2c61NIwTcS2nGYMeM-T7hfhI0BMnnhI0,852
@@ -202,6 +203,6 @@ 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-0.1.11a0.dist-info/METADATA,sha256=t6yN34ubTvc0FxOFA5KhS9o0GJU2zuWyvl2PqeLBxZk,26753
-sarvamai-0.1.11a0.dist-info/WHEEL,sha256=Zb28QaM1gQi8f4VCBhsUklF61CTlNYfs9YAZn-TOGFk,88
-sarvamai-0.1.11a0.dist-info/RECORD,,
+sarvamai-0.1.11a2.dist-info/METADATA,sha256=r4vET9Ft9Lgl9wbRLlh3gR7Hd1LdyaQJjfUKd59n0lY,26753
+sarvamai-0.1.11a2.dist-info/WHEEL,sha256=Zb28QaM1gQi8f4VCBhsUklF61CTlNYfs9YAZn-TOGFk,88
+sarvamai-0.1.11a2.dist-info/RECORD,,

sarvamai/types/audio_codec.py

@@ -1,5 +0,0 @@
-# This file was auto-generated by Fern from our API Definition.
-
-import typing
-
-AudioCodec = typing.Union[typing.Literal["mp3", "linear16", "mulaw", "alaw", "opus", "flac", "aac", "wav"], typing.Any]

sarvamai/types/format.py

@@ -1,5 +0,0 @@
-# This file was auto-generated by Fern from our API Definition.
-
-import typing
-
-Format = typing.Union[typing.Literal["wav", "mp3"], typing.Any]