fix(audio): correct response_format translations type (#1743)

Author: stainless-app[bot]

Diff for: .stats.yml

@@ -1,2 +1,2 @@
 configured_endpoints: 68
-openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai-de1981b64ac229493473670d618500c6362c195f1057eb7de00bd1bc9184fbd5.yml
+openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai-073331021d48db6af646a3552ab0c682efe31b7fb4e59a109ed1ba539f9b89c5.yml

Diff for: api.md

@@ -113,7 +113,7 @@ Methods:
 Types:
 
 ```python
-from openai.types import AudioModel
+from openai.types import AudioModel, AudioResponseFormat
 ```
 
 ## Transcriptions

Diff for: src/openai/resources/audio/transcriptions.py

@@ -8,6 +8,7 @@
 import httpx
 
 from ... import _legacy_response
+from ...types import AudioResponseFormat
 from ..._types import NOT_GIVEN, Body, Query, Headers, NotGiven, FileTypes
 from ..._utils import (
     extract_files,
@@ -22,6 +23,7 @@
 from ..._base_client import make_request_options
 from ...types.audio_model import AudioModel
 from ...types.audio.transcription import Transcription
+from ...types.audio_response_format import AudioResponseFormat
 
 __all__ = ["Transcriptions", "AsyncTranscriptions"]
 
@@ -53,7 +55,7 @@ def create(
         model: Union[str, AudioModel],
         language: str | NotGiven = NOT_GIVEN,
         prompt: str | NotGiven = NOT_GIVEN,
-        response_format: Literal["json", "text", "srt", "verbose_json", "vtt"] | NotGiven = NOT_GIVEN,
+        response_format: AudioResponseFormat | NotGiven = NOT_GIVEN,
         temperature: float | NotGiven = NOT_GIVEN,
         timestamp_granularities: List[Literal["word", "segment"]] | NotGiven = NOT_GIVEN,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
@@ -83,8 +85,8 @@ def create(
               [prompt](https://platform.openai.com/docs/guides/speech-to-text/prompting)
               should match the audio language.
 
-          response_format: The format of the transcript output, in one of these options: `json`, `text`,
-              `srt`, `verbose_json`, or `vtt`.
+          response_format: The format of the output, in one of these options: `json`, `text`, `srt`,
+              `verbose_json`, or `vtt`.
 
           temperature: The sampling temperature, between 0 and 1. Higher values like 0.8 will make the
               output more random, while lower values like 0.2 will make it more focused and
@@ -160,7 +162,7 @@ async def create(
         model: Union[str, AudioModel],
         language: str | NotGiven = NOT_GIVEN,
         prompt: str | NotGiven = NOT_GIVEN,
-        response_format: Literal["json", "text", "srt", "verbose_json", "vtt"] | NotGiven = NOT_GIVEN,
+        response_format: AudioResponseFormat | NotGiven = NOT_GIVEN,
         temperature: float | NotGiven = NOT_GIVEN,
         timestamp_granularities: List[Literal["word", "segment"]] | NotGiven = NOT_GIVEN,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
@@ -190,8 +192,8 @@ async def create(
               [prompt](https://platform.openai.com/docs/guides/speech-to-text/prompting)
               should match the audio language.
 
-          response_format: The format of the transcript output, in one of these options: `json`, `text`,
-              `srt`, `verbose_json`, or `vtt`.
+          response_format: The format of the output, in one of these options: `json`, `text`, `srt`,
+              `verbose_json`, or `vtt`.
 
           temperature: The sampling temperature, between 0 and 1. Higher values like 0.8 will make the
               output more random, while lower values like 0.2 will make it more focused and

Diff for: src/openai/resources/audio/translations.py

@@ -7,6 +7,7 @@
 import httpx
 
 from ... import _legacy_response
+from ...types import AudioResponseFormat
 from ..._types import NOT_GIVEN, Body, Query, Headers, NotGiven, FileTypes
 from ..._utils import (
     extract_files,
@@ -21,6 +22,7 @@
 from ..._base_client import make_request_options
 from ...types.audio_model import AudioModel
 from ...types.audio.translation import Translation
+from ...types.audio_response_format import AudioResponseFormat
 
 __all__ = ["Translations", "AsyncTranslations"]
 
@@ -51,7 +53,7 @@ def create(
         file: FileTypes,
         model: Union[str, AudioModel],
         prompt: str | NotGiven = NOT_GIVEN,
-        response_format: str | NotGiven = NOT_GIVEN,
+        response_format: AudioResponseFormat | NotGiven = NOT_GIVEN,
         temperature: float | NotGiven = NOT_GIVEN,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
@@ -75,8 +77,8 @@ def create(
               [prompt](https://platform.openai.com/docs/guides/speech-to-text/prompting)
               should be in English.
 
-          response_format: The format of the transcript output, in one of these options: `json`, `text`,
-              `srt`, `verbose_json`, or `vtt`.
+          response_format: The format of the output, in one of these options: `json`, `text`, `srt`,
+              `verbose_json`, or `vtt`.
 
           temperature: The sampling temperature, between 0 and 1. Higher values like 0.8 will make the
               output more random, while lower values like 0.2 will make it more focused and
@@ -143,7 +145,7 @@ async def create(
         file: FileTypes,
         model: Union[str, AudioModel],
         prompt: str | NotGiven = NOT_GIVEN,
-        response_format: str | NotGiven = NOT_GIVEN,
+        response_format: AudioResponseFormat | NotGiven = NOT_GIVEN,
         temperature: float | NotGiven = NOT_GIVEN,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
@@ -167,8 +169,8 @@ async def create(
               [prompt](https://platform.openai.com/docs/guides/speech-to-text/prompting)
               should be in English.
 
-          response_format: The format of the transcript output, in one of these options: `json`, `text`,
-              `srt`, `verbose_json`, or `vtt`.
+          response_format: The format of the output, in one of these options: `json`, `text`, `srt`,
+              `verbose_json`, or `vtt`.
 
           temperature: The sampling temperature, between 0 and 1. Higher values like 0.8 will make the
               output more random, while lower values like 0.2 will make it more focused and

Diff for: src/openai/types/__init__.py

@@ -38,6 +38,7 @@
 from .batch_create_params import BatchCreateParams as BatchCreateParams
 from .batch_request_counts import BatchRequestCounts as BatchRequestCounts
 from .upload_create_params import UploadCreateParams as UploadCreateParams
+from .audio_response_format import AudioResponseFormat as AudioResponseFormat
 from .image_generate_params import ImageGenerateParams as ImageGenerateParams
 from .upload_complete_params import UploadCompleteParams as UploadCompleteParams
 from .embedding_create_params import EmbeddingCreateParams as EmbeddingCreateParams

Diff for: src/openai/types/audio/transcription_create_params.py

@@ -7,6 +7,7 @@
 
 from ..._types import FileTypes
 from ..audio_model import AudioModel
+from ..audio_response_format import AudioResponseFormat
 
 __all__ = ["TranscriptionCreateParams"]
 
@@ -41,10 +42,10 @@ class TranscriptionCreateParams(TypedDict, total=False):
     should match the audio language.
     """
 
-    response_format: Literal["json", "text", "srt", "verbose_json", "vtt"]
+    response_format: AudioResponseFormat
     """
-    The format of the transcript output, in one of these options: `json`, `text`,
-    `srt`, `verbose_json`, or `vtt`.
+    The format of the output, in one of these options: `json`, `text`, `srt`,
+    `verbose_json`, or `vtt`.
     """
 
     temperature: float

Diff for: src/openai/types/audio/translation_create_params.py

@@ -7,6 +7,7 @@
 
 from ..._types import FileTypes
 from ..audio_model import AudioModel
+from ..audio_response_format import AudioResponseFormat
 
 __all__ = ["TranslationCreateParams"]
 
@@ -33,10 +34,10 @@ class TranslationCreateParams(TypedDict, total=False):
     should be in English.
     """
 
-    response_format: str
+    response_format: AudioResponseFormat
     """
-    The format of the transcript output, in one of these options: `json`, `text`,
-    `srt`, `verbose_json`, or `vtt`.
+    The format of the output, in one of these options: `json`, `text`, `srt`,
+    `verbose_json`, or `vtt`.
     """
 
     temperature: float

Diff for: src/openai/types/audio_response_format.py

@@ -0,0 +1,7 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing_extensions import Literal, TypeAlias
+
+__all__ = ["AudioResponseFormat"]
+
+AudioResponseFormat: TypeAlias = Literal["json", "text", "srt", "verbose_json", "vtt"]

Diff for: tests/api_resources/audio/test_translations.py

@@ -31,7 +31,7 @@ def test_method_create_with_all_params(self, client: OpenAI) -> None:
             file=b"raw file contents",
             model="whisper-1",
             prompt="prompt",
-            response_format="response_format",
+            response_format="json",
             temperature=0,
         )
         assert_matches_type(Translation, translation, path=["response"])
@@ -80,7 +80,7 @@ async def test_method_create_with_all_params(self, async_client: AsyncOpenAI) ->
             file=b"raw file contents",
             model="whisper-1",
             prompt="prompt",
-            response_format="response_format",
+            response_format="json",
             temperature=0,
         )
         assert_matches_type(Translation, translation, path=["response"])