feat(api): new models for TTS, STT, + new audio features for Realtime (#2232)

Author: stainless-app[bot]

.stats.yml

@@ -1,2 +1,2 @@
-configured_endpoints: 81
-openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai%2Fopenai-b26121d5df6eb5d3032a45a267473798b15fcfec76dd44a3256cf1238be05fa4.yml
+configured_endpoints: 82
+openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai%2Fopenai-c22f59c66aec7914b6ee653d3098d1c1c8c16c180d2a158e819c8ddbf476f74b.yml

api.md

@@ -151,7 +151,11 @@ Types:
 ```python
 from openai.types.audio import (
     Transcription,
+    TranscriptionInclude,
     TranscriptionSegment,
+    TranscriptionStreamEvent,
+    TranscriptionTextDeltaEvent,
+    TranscriptionTextDoneEvent,
     TranscriptionVerbose,
     TranscriptionWord,
     TranscriptionCreateResponse,
@@ -338,7 +342,9 @@ from openai.types.beta.realtime import (
     ConversationItemDeleteEvent,
     ConversationItemDeletedEvent,
     ConversationItemInputAudioTranscriptionCompletedEvent,
+    ConversationItemInputAudioTranscriptionDeltaEvent,
     ConversationItemInputAudioTranscriptionFailedEvent,
+    ConversationItemRetrieveEvent,
     ConversationItemTruncateEvent,
     ConversationItemTruncatedEvent,
     ConversationItemWithReference,
@@ -375,6 +381,8 @@ from openai.types.beta.realtime import (
     SessionCreatedEvent,
     SessionUpdateEvent,
     SessionUpdatedEvent,
+    TranscriptionSessionUpdate,
+    TranscriptionSessionUpdatedEvent,
 )
 ```
 
@@ -390,6 +398,18 @@ Methods:
 
 - <code title="post /realtime/sessions">client.beta.realtime.sessions.<a href="./src/openai/resources/beta/realtime/sessions.py">create</a>(\*\*<a href="src/openai/types/beta/realtime/session_create_params.py">params</a>) -> <a href="./src/openai/types/beta/realtime/session_create_response.py">SessionCreateResponse</a></code>
 
+### TranscriptionSessions
+
+Types:
+
+```python
+from openai.types.beta.realtime import TranscriptionSession
+```
+
+Methods:
+
+- <code title="post /realtime/transcription_sessions">client.beta.realtime.transcription_sessions.<a href="./src/openai/resources/beta/realtime/transcription_sessions.py">create</a>(\*\*<a href="src/openai/types/beta/realtime/transcription_session_create_params.py">params</a>) -> <a href="./src/openai/types/beta/realtime/transcription_session.py">TranscriptionSession</a></code>
+
 ## Assistants
 
 Types:

src/openai/resources/audio/speech.py

@@ -54,6 +54,7 @@ def create(
         input: str,
         model: Union[str, SpeechModel],
         voice: Literal["alloy", "ash", "coral", "echo", "fable", "onyx", "nova", "sage", "shimmer"],
+        instructions: str | NotGiven = NOT_GIVEN,
         response_format: Literal["mp3", "opus", "aac", "flac", "wav", "pcm"] | NotGiven = NOT_GIVEN,
         speed: float | NotGiven = NOT_GIVEN,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
@@ -71,13 +72,16 @@ def create(
 
           model:
               One of the available [TTS models](https://platform.openai.com/docs/models#tts):
-              `tts-1` or `tts-1-hd`
+              `tts-1`, `tts-1-hd` or `gpt-4o-mini-tts`.
 
           voice: The voice to use when generating the audio. Supported voices are `alloy`, `ash`,
               `coral`, `echo`, `fable`, `onyx`, `nova`, `sage` and `shimmer`. Previews of the
               voices are available in the
               [Text to speech guide](https://platform.openai.com/docs/guides/text-to-speech#voice-options).
 
+          instructions: Control the voice of your generated audio with additional instructions. Does not
+              work with `tts-1` or `tts-1-hd`.
+
           response_format: The format to audio in. Supported formats are `mp3`, `opus`, `aac`, `flac`,
               `wav`, and `pcm`.
 
@@ -100,6 +104,7 @@ def create(
                     "input": input,
                     "model": model,
                     "voice": voice,
+                    "instructions": instructions,
                     "response_format": response_format,
                     "speed": speed,
                 },
@@ -138,6 +143,7 @@ async def create(
         input: str,
         model: Union[str, SpeechModel],
         voice: Literal["alloy", "ash", "coral", "echo", "fable", "onyx", "nova", "sage", "shimmer"],
+        instructions: str | NotGiven = NOT_GIVEN,
         response_format: Literal["mp3", "opus", "aac", "flac", "wav", "pcm"] | NotGiven = NOT_GIVEN,
         speed: float | NotGiven = NOT_GIVEN,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
@@ -155,13 +161,16 @@ async def create(
 
           model:
               One of the available [TTS models](https://platform.openai.com/docs/models#tts):
-              `tts-1` or `tts-1-hd`
+              `tts-1`, `tts-1-hd` or `gpt-4o-mini-tts`.
 
           voice: The voice to use when generating the audio. Supported voices are `alloy`, `ash`,
               `coral`, `echo`, `fable`, `onyx`, `nova`, `sage` and `shimmer`. Previews of the
               voices are available in the
               [Text to speech guide](https://platform.openai.com/docs/guides/text-to-speech#voice-options).
 
+          instructions: Control the voice of your generated audio with additional instructions. Does not
+              work with `tts-1` or `tts-1-hd`.
+
           response_format: The format to audio in. Supported formats are `mp3`, `opus`, `aac`, `flac`,
               `wav`, and `pcm`.
 
@@ -184,6 +193,7 @@ async def create(
                     "input": input,
                     "model": model,
                     "voice": voice,
+                    "instructions": instructions,
                     "response_format": response_format,
                     "speed": speed,
                 },