sarvamai 0.1.16a0__py3-none-any.whl → 0.1.17a0__py3-none-any.whl

sarvamai/__init__.py

@@ -174,6 +174,7 @@ from .requests import (
     TranslationResponseParams,
     TransliterationResponseParams,
 )
+from .speech_to_text import SpeechToTextTranscribeRequestInputAudioCodec, SpeechToTextTranslateRequestInputAudioCodec
 from .speech_to_text_streaming import (
     SpeechToTextStreamingHighVadSensitivity,
     SpeechToTextStreamingLanguageCode,
@@ -301,12 +302,14 @@ __all__ = [
     "SpeechToTextStreamingResponse",
     "SpeechToTextStreamingResponseParams",
     "SpeechToTextStreamingVadSignals",
+    "SpeechToTextTranscribeRequestInputAudioCodec",
     "SpeechToTextTranscriptionData",
     "SpeechToTextTranscriptionDataParams",
     "SpeechToTextTranslateJobParameters",
     "SpeechToTextTranslateJobParametersParams",
     "SpeechToTextTranslateLanguage",
     "SpeechToTextTranslateModel",
+    "SpeechToTextTranslateRequestInputAudioCodec",
     "SpeechToTextTranslateResponse",
     "SpeechToTextTranslateResponseData",
     "SpeechToTextTranslateResponseDataParams",

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/0.1.16a0",
+            "User-Agent": "sarvamai/0.1.17a0",
             "X-Fern-Language": "Python",
             "X-Fern-SDK-Name": "sarvamai",
-            "X-Fern-SDK-Version": "0.1.16a0",
+            "X-Fern-SDK-Version": "0.1.17a0",
             **(self.get_custom_headers() or {}),
         }
         headers["api-subscription-key"] = self.api_subscription_key

sarvamai/speech_to_text/__init__.py

@@ -2,3 +2,6 @@
 
 # isort: skip_file
 
+from .types import SpeechToTextTranscribeRequestInputAudioCodec, SpeechToTextTranslateRequestInputAudioCodec
+
+__all__ = ["SpeechToTextTranscribeRequestInputAudioCodec", "SpeechToTextTranslateRequestInputAudioCodec"]

sarvamai/speech_to_text/client.py

@@ -11,6 +11,8 @@ from ..types.speech_to_text_response import SpeechToTextResponse
 from ..types.speech_to_text_translate_model import SpeechToTextTranslateModel
 from ..types.speech_to_text_translate_response import SpeechToTextTranslateResponse
 from .raw_client import AsyncRawSpeechToTextClient, RawSpeechToTextClient
+from .types.speech_to_text_transcribe_request_input_audio_codec import SpeechToTextTranscribeRequestInputAudioCodec
+from .types.speech_to_text_translate_request_input_audio_codec import SpeechToTextTranslateRequestInputAudioCodec
 
 # this is used as the default value for optional parameters
 OMIT = typing.cast(typing.Any, ...)
@@ -37,6 +39,7 @@ class SpeechToTextClient:
         file: core.File,
         model: typing.Optional[SpeechToTextModel] = OMIT,
         language_code: typing.Optional[SpeechToTextLanguage] = OMIT,
+        input_audio_codec: typing.Optional[SpeechToTextTranscribeRequestInputAudioCodec] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> SpeechToTextResponse:
         """
@@ -68,6 +71,9 @@ class SpeechToTextClient:
              For the `saarika:v2.5` model, it is optional.
             `unknown`: Use this when the language is not known; the API will detect it automatically.
 
+        input_audio_codec : typing.Optional[SpeechToTextTranscribeRequestInputAudioCodec]
+            Input Audio codec/format of the input file. PCM files are supported only at 16kHz sample rate.
+
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
@@ -86,7 +92,11 @@ class SpeechToTextClient:
         client.speech_to_text.transcribe()
         """
         _response = self._raw_client.transcribe(
-            file=file, model=model, language_code=language_code, request_options=request_options
+            file=file,
+            model=model,
+            language_code=language_code,
+            input_audio_codec=input_audio_codec,
+            request_options=request_options,
         )
         return _response.data
 
@@ -96,6 +106,7 @@ class SpeechToTextClient:
         file: core.File,
         prompt: typing.Optional[str] = OMIT,
         model: typing.Optional[SpeechToTextTranslateModel] = OMIT,
+        input_audio_codec: typing.Optional[SpeechToTextTranslateRequestInputAudioCodec] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> SpeechToTextTranslateResponse:
         """
@@ -119,11 +130,14 @@ class SpeechToTextClient:
             See core.File for more documentation
 
         prompt : typing.Optional[str]
-            Conversation context can be passed as a prompt to boost model accuracy. However, the current system is at an experimentation stage and doesn’t match the prompt performance of large language models.
+            Conversation context can be passed as a prompt to boost model accuracy. However, the current system is at an experimentation stage and doesn't match the prompt performance of large language models.
 
         model : typing.Optional[SpeechToTextTranslateModel]
             Model to be used for converting speech to text in target language
 
+        input_audio_codec : typing.Optional[SpeechToTextTranslateRequestInputAudioCodec]
+            Audio codec/format of the input file. Our API automatically detects all codec formats, but for PCM files specifically (pcm_s16le, pcm_l16, pcm_raw), you must pass this parameter. PCM files are supported only at 16kHz sample rate.
+
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
@@ -141,7 +155,9 @@ class SpeechToTextClient:
         )
         client.speech_to_text.translate()
         """
-        _response = self._raw_client.translate(file=file, prompt=prompt, model=model, request_options=request_options)
+        _response = self._raw_client.translate(
+            file=file, prompt=prompt, model=model, input_audio_codec=input_audio_codec, request_options=request_options
+        )
         return _response.data
 
 
@@ -166,6 +182,7 @@ class AsyncSpeechToTextClient:
         file: core.File,
         model: typing.Optional[SpeechToTextModel] = OMIT,
         language_code: typing.Optional[SpeechToTextLanguage] = OMIT,
+        input_audio_codec: typing.Optional[SpeechToTextTranscribeRequestInputAudioCodec] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> SpeechToTextResponse:
         """
@@ -197,6 +214,9 @@ class AsyncSpeechToTextClient:
              For the `saarika:v2.5` model, it is optional.
             `unknown`: Use this when the language is not known; the API will detect it automatically.
 
+        input_audio_codec : typing.Optional[SpeechToTextTranscribeRequestInputAudioCodec]
+            Input Audio codec/format of the input file. PCM files are supported only at 16kHz sample rate.
+
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
@@ -223,7 +243,11 @@ class AsyncSpeechToTextClient:
         asyncio.run(main())
         """
         _response = await self._raw_client.transcribe(
-            file=file, model=model, language_code=language_code, request_options=request_options
+            file=file,
+            model=model,
+            language_code=language_code,
+            input_audio_codec=input_audio_codec,
+            request_options=request_options,
         )
         return _response.data
 
@@ -233,6 +257,7 @@ class AsyncSpeechToTextClient:
         file: core.File,
         prompt: typing.Optional[str] = OMIT,
         model: typing.Optional[SpeechToTextTranslateModel] = OMIT,
+        input_audio_codec: typing.Optional[SpeechToTextTranslateRequestInputAudioCodec] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> SpeechToTextTranslateResponse:
         """
@@ -256,11 +281,14 @@ class AsyncSpeechToTextClient:
             See core.File for more documentation
 
         prompt : typing.Optional[str]
-            Conversation context can be passed as a prompt to boost model accuracy. However, the current system is at an experimentation stage and doesn’t match the prompt performance of large language models.
+            Conversation context can be passed as a prompt to boost model accuracy. However, the current system is at an experimentation stage and doesn't match the prompt performance of large language models.
 
         model : typing.Optional[SpeechToTextTranslateModel]
             Model to be used for converting speech to text in target language
 
+        input_audio_codec : typing.Optional[SpeechToTextTranslateRequestInputAudioCodec]
+            Audio codec/format of the input file. Our API automatically detects all codec formats, but for PCM files specifically (pcm_s16le, pcm_l16, pcm_raw), you must pass this parameter. PCM files are supported only at 16kHz sample rate.
+
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
@@ -287,6 +315,6 @@ class AsyncSpeechToTextClient:
         asyncio.run(main())
         """
         _response = await self._raw_client.translate(
-            file=file, prompt=prompt, model=model, request_options=request_options
+            file=file, prompt=prompt, model=model, input_audio_codec=input_audio_codec, request_options=request_options
         )
         return _response.data

sarvamai/speech_to_text/raw_client.py

@@ -20,6 +20,8 @@ from ..types.speech_to_text_model import SpeechToTextModel
 from ..types.speech_to_text_response import SpeechToTextResponse
 from ..types.speech_to_text_translate_model import SpeechToTextTranslateModel
 from ..types.speech_to_text_translate_response import SpeechToTextTranslateResponse
+from .types.speech_to_text_transcribe_request_input_audio_codec import SpeechToTextTranscribeRequestInputAudioCodec
+from .types.speech_to_text_translate_request_input_audio_codec import SpeechToTextTranslateRequestInputAudioCodec
 
 # this is used as the default value for optional parameters
 OMIT = typing.cast(typing.Any, ...)
@@ -35,6 +37,7 @@ class RawSpeechToTextClient:
         file: core.File,
         model: typing.Optional[SpeechToTextModel] = OMIT,
         language_code: typing.Optional[SpeechToTextLanguage] = OMIT,
+        input_audio_codec: typing.Optional[SpeechToTextTranscribeRequestInputAudioCodec] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> HttpResponse[SpeechToTextResponse]:
         """
@@ -66,6 +69,9 @@ class RawSpeechToTextClient:
              For the `saarika:v2.5` model, it is optional.
             `unknown`: Use this when the language is not known; the API will detect it automatically.
 
+        input_audio_codec : typing.Optional[SpeechToTextTranscribeRequestInputAudioCodec]
+            Input Audio codec/format of the input file. PCM files are supported only at 16kHz sample rate.
+
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
@@ -81,6 +87,7 @@ class RawSpeechToTextClient:
             data={
                 "model": model,
                 "language_code": language_code,
+                "input_audio_codec": input_audio_codec,
             },
             files={
                 "file": file,
@@ -176,6 +183,7 @@ class RawSpeechToTextClient:
         file: core.File,
         prompt: typing.Optional[str] = OMIT,
         model: typing.Optional[SpeechToTextTranslateModel] = OMIT,
+        input_audio_codec: typing.Optional[SpeechToTextTranslateRequestInputAudioCodec] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> HttpResponse[SpeechToTextTranslateResponse]:
         """
@@ -199,11 +207,14 @@ class RawSpeechToTextClient:
             See core.File for more documentation
 
         prompt : typing.Optional[str]
-            Conversation context can be passed as a prompt to boost model accuracy. However, the current system is at an experimentation stage and doesn’t match the prompt performance of large language models.
+            Conversation context can be passed as a prompt to boost model accuracy. However, the current system is at an experimentation stage and doesn't match the prompt performance of large language models.
 
         model : typing.Optional[SpeechToTextTranslateModel]
             Model to be used for converting speech to text in target language
 
+        input_audio_codec : typing.Optional[SpeechToTextTranslateRequestInputAudioCodec]
+            Audio codec/format of the input file. Our API automatically detects all codec formats, but for PCM files specifically (pcm_s16le, pcm_l16, pcm_raw), you must pass this parameter. PCM files are supported only at 16kHz sample rate.
+
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
@@ -219,6 +230,7 @@ class RawSpeechToTextClient:
             data={
                 "prompt": prompt,
                 "model": model,
+                "input_audio_codec": input_audio_codec,
             },
             files={
                 "file": file,
@@ -319,6 +331,7 @@ class AsyncRawSpeechToTextClient:
         file: core.File,
         model: typing.Optional[SpeechToTextModel] = OMIT,
         language_code: typing.Optional[SpeechToTextLanguage] = OMIT,
+        input_audio_codec: typing.Optional[SpeechToTextTranscribeRequestInputAudioCodec] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> AsyncHttpResponse[SpeechToTextResponse]:
         """
@@ -350,6 +363,9 @@ class AsyncRawSpeechToTextClient:
              For the `saarika:v2.5` model, it is optional.
             `unknown`: Use this when the language is not known; the API will detect it automatically.
 
+        input_audio_codec : typing.Optional[SpeechToTextTranscribeRequestInputAudioCodec]
+            Input Audio codec/format of the input file. PCM files are supported only at 16kHz sample rate.
+
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
@@ -365,6 +381,7 @@ class AsyncRawSpeechToTextClient:
             data={
                 "model": model,
                 "language_code": language_code,
+                "input_audio_codec": input_audio_codec,
             },
             files={
                 "file": file,
@@ -460,6 +477,7 @@ class AsyncRawSpeechToTextClient:
         file: core.File,
         prompt: typing.Optional[str] = OMIT,
         model: typing.Optional[SpeechToTextTranslateModel] = OMIT,
+        input_audio_codec: typing.Optional[SpeechToTextTranslateRequestInputAudioCodec] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> AsyncHttpResponse[SpeechToTextTranslateResponse]:
         """
@@ -483,11 +501,14 @@ class AsyncRawSpeechToTextClient:
             See core.File for more documentation
 
         prompt : typing.Optional[str]
-            Conversation context can be passed as a prompt to boost model accuracy. However, the current system is at an experimentation stage and doesn’t match the prompt performance of large language models.
+            Conversation context can be passed as a prompt to boost model accuracy. However, the current system is at an experimentation stage and doesn't match the prompt performance of large language models.
 
         model : typing.Optional[SpeechToTextTranslateModel]
             Model to be used for converting speech to text in target language
 
+        input_audio_codec : typing.Optional[SpeechToTextTranslateRequestInputAudioCodec]
+            Audio codec/format of the input file. Our API automatically detects all codec formats, but for PCM files specifically (pcm_s16le, pcm_l16, pcm_raw), you must pass this parameter. PCM files are supported only at 16kHz sample rate.
+
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
@@ -503,6 +524,7 @@ class AsyncRawSpeechToTextClient:
             data={
                 "prompt": prompt,
                 "model": model,
+                "input_audio_codec": input_audio_codec,
             },
             files={
                 "file": file,

sarvamai/speech_to_text/types/__init__.py

@@ -0,0 +1,8 @@
+# This file was auto-generated by Fern from our API Definition.
+
+# isort: skip_file
+
+from .speech_to_text_transcribe_request_input_audio_codec import SpeechToTextTranscribeRequestInputAudioCodec
+from .speech_to_text_translate_request_input_audio_codec import SpeechToTextTranslateRequestInputAudioCodec
+
+__all__ = ["SpeechToTextTranscribeRequestInputAudioCodec", "SpeechToTextTranslateRequestInputAudioCodec"]

sarvamai/speech_to_text/types/speech_to_text_transcribe_request_input_audio_codec.py

@@ -0,0 +1,33 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+SpeechToTextTranscribeRequestInputAudioCodec = typing.Union[
+    typing.Literal[
+        "wav",
+        "x-wav",
+        "wave",
+        "mp3",
+        "mpeg",
+        "mpeg3",
+        "x-mp3",
+        "x-mpeg-3",
+        "aac",
+        "x-aac",
+        "aiff",
+        "x-aiff",
+        "ogg",
+        "opus",
+        "flac",
+        "x-flac",
+        "mp4",
+        "x-m4a",
+        "amr",
+        "x-ms-wma",
+        "webm",
+        "pcm_s16le",
+        "pcm_l16",
+        "pcm_raw",
+    ],
+    typing.Any,
+]

sarvamai/speech_to_text/types/speech_to_text_translate_request_input_audio_codec.py

@@ -0,0 +1,33 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+SpeechToTextTranslateRequestInputAudioCodec = typing.Union[
+    typing.Literal[
+        "wav",
+        "x-wav",
+        "wave",
+        "mp3",
+        "mpeg",
+        "mpeg3",
+        "x-mp3",
+        "x-mpeg-3",
+        "aac",
+        "x-aac",
+        "aiff",
+        "x-aiff",
+        "ogg",
+        "opus",
+        "flac",
+        "x-flac",
+        "mp4",
+        "x-m4a",
+        "amr",
+        "x-ms-wma",
+        "webm",
+        "pcm_s16le",
+        "pcm_l16",
+        "pcm_raw",
+    ],
+    typing.Any,
+]

{sarvamai-0.1.16a0.dist-info → sarvamai-0.1.17a0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sarvamai
-Version: 0.1.16a0
+Version: 0.1.17a0
 Summary: 
 Requires-Python: >=3.8,<4.0
 Classifier: Intended Audience :: Developers

{sarvamai-0.1.16a0.dist-info → sarvamai-0.1.17a0.dist-info}/RECORD

@@ -1,11 +1,11 @@
-sarvamai/__init__.py,sha256=o9bK5DhrKYzmhZbOvKNNl-Fcif9og4eynOp47Dna4vY,10650
+sarvamai/__init__.py,sha256=_AvjD4uJaddFuA7gBmWXoGqU9JEo9mHpVq-cqk47jws,10871
 sarvamai/chat/__init__.py,sha256=_VhToAyIt_5axN6CLJwtxg3-CO7THa_23pbUzqhXJa4,85
 sarvamai/chat/client.py,sha256=xOSj83Gr6Q7eY2qUeATiuXYQqBqWqSCQlIEopK5fKus,11022
 sarvamai/chat/raw_client.py,sha256=A2kRuZcVWlJhyYCD7YKgqNkZEp3cYa1731KhRkhirU0,17885
 sarvamai/client.py,sha256=J30X_os1lPf8Wml0KDFEf6p8VGHhgF_lf3nw1T2D3qo,8207
 sarvamai/core/__init__.py,sha256=YE2CtXeASe1RAbaI39twKWYKCuT4tW5is9HWHhJjR_g,1653
 sarvamai/core/api_error.py,sha256=44vPoTyWN59gonCIZMdzw7M1uspygiLnr3GNFOoVL2Q,614
-sarvamai/core/client_wrapper.py,sha256=ya3guU64Ap76A6U2WEq0HFnOz6Ds_qLLmIqk0NgoRWE,2570
+sarvamai/core/client_wrapper.py,sha256=OJ4aojUSt1w1Qnvo0h5xwdvbFZKBybSJqwWw9tvkbVM,2570
 sarvamai/core/datetime_utils.py,sha256=nBys2IsYrhPdszxGKCNRPSOCwa-5DWOHG95FB8G9PKo,1047
 sarvamai/core/events.py,sha256=HvKBdSoYcFetk7cgNXb7FxuY-FtY8NtUhZIN7mGVx8U,1159
 sarvamai/core/file.py,sha256=d4NNbX8XvXP32z8KpK2Xovv33nFfruIrpz0QWxlgpZk,2663
@@ -83,9 +83,12 @@ sarvamai/requests/timestamps_model.py,sha256=LkaYC5poipZe4r5-7Tp_Y058RXYWXwocn46
 sarvamai/requests/transcription_metrics.py,sha256=FDclX2Z9Z3azrDXxtZW8xbkxxWMZQXEbaYMMheAAAdk,321
 sarvamai/requests/translation_response.py,sha256=8iwQeZB1purHY757bIQI-n9QeVRBItaAVcBJ_la-k1Y,414
 sarvamai/requests/transliteration_response.py,sha256=KqRkqnegLmt7LjdVxjRePX6RoqaLm64KFGZ6q7mXyfw,426
-sarvamai/speech_to_text/__init__.py,sha256=_VhToAyIt_5axN6CLJwtxg3-CO7THa_23pbUzqhXJa4,85
-sarvamai/speech_to_text/client.py,sha256=v9p6bqRbA9h3f4wuylPTVDwetG7MCOlEtmrqW18KHCU,10857
-sarvamai/speech_to_text/raw_client.py,sha256=IhflpQX1P2kvZw8L2qbneMbRs5gtr9ye6ebvEK-hn3A,25185
+sarvamai/speech_to_text/__init__.py,sha256=U3RTBb6u3QwsV3fyh-RX_HQ3AI2BD0i-Y8ec3gqTjdI,301
+sarvamai/speech_to_text/client.py,sha256=Z_MZH-W70BbNYkax_CV3ZIhgxT0d44rnXtN3yLTMIzc,12797
+sarvamai/speech_to_text/raw_client.py,sha256=kHU4ngENfwqrKBZq0GZwAlYONPWf3ssH3hQvJhEyEXg,27081
+sarvamai/speech_to_text/types/__init__.py,sha256=HIUDEN0mwbyXM15kMPzNydaxB9FEY_Fjzj5RsMFyef4,410
+sarvamai/speech_to_text/types/speech_to_text_transcribe_request_input_audio_codec.py,sha256=FRi5ofYQFd4TmDbAqZ0Dx2mL0F7FRovwJGn1OI8N4q0,592
+sarvamai/speech_to_text/types/speech_to_text_translate_request_input_audio_codec.py,sha256=q5ZnWwicyFtdOX9EMDUkHipSIIAMstCVYyjesminEWM,591
 sarvamai/speech_to_text_job/__init__.py,sha256=_VhToAyIt_5axN6CLJwtxg3-CO7THa_23pbUzqhXJa4,85
 sarvamai/speech_to_text_job/client.py,sha256=WSGBJxYcNxl77Zd1X6VVWjg4zshqecXf6WCyhfLXVlI,18007
 sarvamai/speech_to_text_job/job.py,sha256=K8HOmwrYd6l82-MZfWDBmNkZeeERyg9YOihnFfvl-Js,15021
@@ -210,6 +213,6 @@ 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-0.1.16a0.dist-info/METADATA,sha256=kJFllmgiNTGqzAiFZ0nntIXPASSibBS3sRNTIcdaWhk,26753
-sarvamai-0.1.16a0.dist-info/WHEEL,sha256=Zb28QaM1gQi8f4VCBhsUklF61CTlNYfs9YAZn-TOGFk,88
-sarvamai-0.1.16a0.dist-info/RECORD,,
+sarvamai-0.1.17a0.dist-info/METADATA,sha256=Frl6Uw6tD1u701FFbq4j6anJgmBbbNDmZr_ZNZlZfh8,26753
+sarvamai-0.1.17a0.dist-info/WHEEL,sha256=Zb28QaM1gQi8f4VCBhsUklF61CTlNYfs9YAZn-TOGFk,88
+sarvamai-0.1.17a0.dist-info/RECORD,,