release: 1.68.0

.gitignore

@@ -14,3 +14,7 @@ dist
 .envrc
 codegen.log
 Brewfile.lock.json
+
+.DS_Store
+
+examples/*.mp3

.release-please-manifest.json

@@ -1,3 +1,3 @@
 {
-  ".": "1.67.0"
+  ".": "1.68.0"
 }

.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

CHANGELOG.md

@@ -1,5 +1,14 @@
 # Changelog
 
+## 1.68.0 (2025-03-20)
+
+Full Changelog: [v1.67.0...v1.68.0](https://github.com/openai/openai-python/compare/v1.67.0...v1.68.0)
+
+### Features
+
+* add audio helpers ([423655c](https://github.com/openai/openai-python/commit/423655ca9077cfd258f1e52f6eb386fc8307fa5f))
+* **api:** new models for TTS, STT, + new audio features for Realtime ([#2232](https://github.com/openai/openai-python/issues/2232)) ([ab5192d](https://github.com/openai/openai-python/commit/ab5192d0a7b417ade622ec94dd48f86beb90692c))
+
 ## 1.67.0 (2025-03-19)
 
 Full Changelog: [v1.66.5...v1.67.0](https://github.com/openai/openai-python/compare/v1.66.5...v1.67.0)

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:

examples/audio.py

@@ -1,6 +1,5 @@
 #!/usr/bin/env rye run python
 
-import time
 from pathlib import Path
 
 from openai import OpenAI
@@ -12,8 +11,6 @@
 
 
 def main() -> None:
-    stream_to_speakers()
-
     # Create text-to-speech audio file
     with openai.audio.speech.with_streaming_response.create(
         model="tts-1",
@@ -37,28 +34,5 @@ def main() -> None:
     print(translation.text)
 
 
-def stream_to_speakers() -> None:
-    import pyaudio
-
-    player_stream = pyaudio.PyAudio().open(format=pyaudio.paInt16, channels=1, rate=24000, output=True)
-
-    start_time = time.time()
-
-    with openai.audio.speech.with_streaming_response.create(
-        model="tts-1",
-        voice="alloy",
-        response_format="pcm",  # similar to WAV, but without a header chunk at the start.
-        input="""I see skies of blue and clouds of white
-                The bright blessed days, the dark sacred nights
-                And I think to myself
-                What a wonderful world""",
-    ) as response:
-        print(f"Time to first byte: {int((time.time() - start_time) * 1000)}ms")
-        for chunk in response.iter_bytes(chunk_size=1024):
-            player_stream.write(chunk)
-
-    print(f"Done in {int((time.time() - start_time) * 1000)}ms.")
-
-
 if __name__ == "__main__":
     main()

examples/speech_to_text.py

@@ -0,0 +1,25 @@
+#!/usr/bin/env rye run python
+
+import asyncio
+
+from openai import AsyncOpenAI
+from openai.helpers import Microphone
+
+# gets OPENAI_API_KEY from your environment variables
+openai = AsyncOpenAI()
+
+
+async def main() -> None:
+    print("Recording for the next 10 seconds...")
+    recording = await Microphone(timeout=10).record()
+    print("Recording complete")
+    transcription = await openai.audio.transcriptions.create(
+        model="whisper-1",
+        file=recording,
+    )
+
+    print(transcription.text)
+
+
+if __name__ == "__main__":
+    asyncio.run(main())

examples/text_to_speech.py

@@ -0,0 +1,31 @@
+#!/usr/bin/env rye run python
+
+import time
+import asyncio
+
+from openai import AsyncOpenAI
+from openai.helpers import LocalAudioPlayer
+
+# gets OPENAI_API_KEY from your environment variables
+openai = AsyncOpenAI()
+
+
+async def main() -> None:
+    start_time = time.time()
+
+    async with openai.audio.speech.with_streaming_response.create(
+        model="tts-1",
+        voice="alloy",
+        response_format="pcm",  # similar to WAV, but without a header chunk at the start.
+        input="""I see skies of blue and clouds of white
+                The bright blessed days, the dark sacred nights
+                And I think to myself
+                What a wonderful world""",
+    ) as response:
+        print(f"Time to first byte: {int((time.time() - start_time) * 1000)}ms")
+        await LocalAudioPlayer().play(response)
+        print(f"Time to play: {int((time.time() - start_time) * 1000)}ms")
+
+
+if __name__ == "__main__":
+    asyncio.run(main())

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "openai"
-version = "1.67.0"
+version = "1.68.0"
 description = "The official Python library for the openai API"
 dynamic = ["readme"]
 license = "Apache-2.0"
@@ -16,6 +16,8 @@ dependencies = [
     "sniffio",
     "tqdm > 4",
     "jiter>=0.4.0, <1",
+    "sounddevice>=0.5.1",
+    "numpy>=2.0.2",
 ]
 requires-python = ">= 3.8"
 classifiers = [

requirements-dev.lock

@@ -33,6 +33,7 @@ certifi==2023.7.22
     # via requests
 cffi==1.16.0
     # via cryptography
+    # via sounddevice
 charset-normalizer==3.3.2
     # via requests
 click==8.1.7
@@ -92,7 +93,7 @@ nest-asyncio==1.6.0
 nodeenv==1.8.0
     # via pyright
 nox==2023.4.22
-numpy==1.26.3
+numpy==2.0.2
     # via openai
     # via pandas
     # via pandas-stubs
@@ -102,7 +103,7 @@ packaging==23.2
     # via black
     # via nox
     # via pytest
-pandas==2.1.4
+pandas==2.2.3
     # via openai
 pandas-stubs==2.1.4.231227
     # via openai
@@ -154,6 +155,8 @@ sniffio==1.3.0
     # via trio
 sortedcontainers==2.4.0
     # via trio
+sounddevice==0.5.1
+    # via openai
 time-machine==2.9.0
 toml==0.10.2
     # via inline-snapshot

requirements.lock

@@ -18,6 +18,8 @@ anyio==4.1.0
 certifi==2023.7.22
     # via httpcore
     # via httpx
+cffi==1.17.1
+    # via sounddevice
 distro==1.8.0
     # via openai
 exceptiongroup==1.2.2
@@ -41,6 +43,8 @@ pandas==2.2.3
     # via openai
 pandas-stubs==2.2.2.240807
     # via openai
+pycparser==2.22
+    # via cffi
 pydantic==2.10.3
     # via openai
 pydantic-core==2.27.1
@@ -54,6 +58,8 @@ six==1.16.0
 sniffio==1.3.0
     # via anyio
     # via openai
+sounddevice==0.5.1
+    # via openai
 tqdm==4.66.5
     # via openai
 types-pytz==2024.2.0.20241003

src/openai/_version.py

@@ -1,4 +1,4 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
 __title__ = "openai"
-__version__ = "1.67.0"  # x-release-please-version
+__version__ = "1.68.0"  # x-release-please-version

src/openai/helpers.py

@@ -0,0 +1,4 @@
+from .helpers.microphone import Microphone
+from .helpers.local_audio_player import LocalAudioPlayer
+
+__all__ = ["LocalAudioPlayer", "Microphone"]

src/openai/helpers/__init__.py

@@ -0,0 +1,4 @@
+from .microphone import Microphone
+from .local_audio_player import LocalAudioPlayer
+
+__all__ = ["Microphone", "LocalAudioPlayer"]