feat!: remove the extras module and the [extras] install option

.github/workflows/test.yml

@@ -43,7 +43,6 @@ jobs:
           python-version: ${{ matrix.py }}
       - name: Setup test suite
         run: |
-          sudo apt-get update && sudo apt-get install -y portaudio19-dev
           python_version="${{ matrix.py }}"
           python_version="${python_version/./}"
           tox -f "py$python_version" -vvvv --notest

CLAUDE.md

@@ -343,7 +343,7 @@ async def main():
 asyncio.run(main())
 ```
 
-**Microphone streaming** (both clients): `pip install "assemblyai[extras]"` for `pyaudio`, then pass `aai.extras.MicrophoneStream(sample_rate=16000)` to `client.stream(...)`.
+**Microphone streaming**: the SDK does not capture audio — supply 16-bit PCM chunks yourself (e.g. via `pyaudio` or `sounddevice`) and pass the iterable to `client.stream(...)`.
 
 **Voice-agent tuning** (knobs that matter most when building a voice agent):
 ```python
@@ -404,7 +404,7 @@ async with AsyncStreamingClient(StreamingClientOptions(token=token_from_server))
 ```
 
 **Other gotchas**:
-- Don't pass `aai.extras.stream_file(...)` to `AsyncStreamingClient.stream()` — it uses blocking `time.sleep` and starves the read task. Use an `async def` generator with `await asyncio.sleep(...)` instead.
+- Don't feed `AsyncStreamingClient.stream()` from a sync generator that blocks (e.g. `time.sleep` pacing) — it starves the read task. Use an `async def` generator with `await asyncio.sleep(...)` instead.
 - `format_turns=True` enables punctuation/casing on confirmed end-of-turns. Toggle mid-session via `client.set_params(StreamingSessionParameters(format_turns=True))`.
 - `AsyncStreamingClient` used as `async with` calls `disconnect(terminate=True)` on normal block exit and `disconnect(terminate=False)` on exception — no explicit `disconnect()` needed inside the block.
 
@@ -414,14 +414,14 @@ async with AsyncStreamingClient(StreamingClientOptions(token=token_from_server))
 - **`speech_models` takes a list** with fallback ordering: `["universal-3-5-pro", "universal-2"]`
 - **PII redaction uses `set_redact_pii()`**, not a constructor parameter
 - **Streaming v3 lives in its own module**: `assemblyai.streaming.v3` (there is no other streaming API in this SDK). See the "Streaming (real-time)" section above.
-- **Microphone streaming needs extras**: `pip install "assemblyai[extras]"` for `pyaudio`
+- **The SDK does not capture microphone audio**: bring your own capture (e.g. `pyaudio`, `sounddevice`) and stream the PCM chunks
 - **`transcribe_async()` returns a `concurrent.futures.Future`**, not an asyncio coroutine. In asyncio code use `aai.AsyncTranscriber` (see "Asyncio transcription" above) — or `aai.AsyncSyncTranscriber` for the sync API
 - **Timestamps are in milliseconds** throughout the SDK
 - **Minimum Python**: 3.8+
 
 ## Dependencies
 
-`httpx`, `pydantic`, `typing-extensions`, `websockets`. Optional: `pyaudio` via `[extras]`.
+`httpx`, `pydantic`, `typing-extensions`, `websockets`.
 
 ## Docs
 

README.md

@@ -1017,12 +1017,20 @@ Real-time speech-to-text via WebSocket against the `universal-3-5-pro` model. Th
   <summary>Stream a local file (sync)</summary>
 
 ```python
-import assemblyai as aai
+import time
+
 from assemblyai.streaming.v3 import (
     BeginEvent, StreamingClient, StreamingClientOptions, StreamingError,
     StreamingEvents, StreamingParameters, TerminationEvent, TurnEvent,
 )
 
+def stream_file(path: str, sample_rate: int, chunk_duration: float = 0.3):
+    bytes_per_chunk = int(sample_rate * chunk_duration) * 2
+    with open(path, "rb") as f:
+        while chunk := f.read(bytes_per_chunk):
+            yield chunk
+            time.sleep(chunk_duration)
+
 def on_begin(client, event: BeginEvent):
     print(f"Session started: {event.id}")
 
@@ -1045,37 +1053,7 @@ client.connect(StreamingParameters(
     sample_rate=16000, speech_model="universal-3-5-pro",
 ))
 try:
-    client.stream(aai.extras.stream_file(filepath="audio.wav", sample_rate=16000))
-finally:
-    client.disconnect(terminate=True)
-```
-
-</details>
-
-<details>
-  <summary>Stream your microphone (sync)</summary>
-
-`MicrophoneStream` requires PyAudio:
-
-```bash
-pip install -U "assemblyai[extras]"
-```
-
-```python
-import assemblyai as aai
-from assemblyai.streaming.v3 import (
-    StreamingClient, StreamingClientOptions, StreamingEvents, StreamingParameters,
-)
-
-def on_turn(client, event):
-    print(f"{event.transcript} (end_of_turn={event.end_of_turn})")
-
-client = StreamingClient(StreamingClientOptions(api_key="<YOUR_API_KEY>"))
-client.on(StreamingEvents.Turn, on_turn)
-client.connect(StreamingParameters(sample_rate=16000, speech_model="universal-3-5-pro"))
-
-try:
-    client.stream(aai.extras.MicrophoneStream(sample_rate=16000))
+    client.stream(stream_file("audio.wav", sample_rate=16000))
 finally:
     client.disconnect(terminate=True)
 ```
@@ -1149,7 +1127,7 @@ See [`examples/streaming_dual_channel.py`](./examples/streaming_dual_channel.py)
 <details>
   <summary>Stream a local file (async)</summary>
 
-`AsyncStreamingClient` mirrors `StreamingClient` with async methods. It's safe to use as an async context manager — `disconnect()` runs on block exit even if user code raises. Don't pass `extras.stream_file` directly (it uses blocking `time.sleep`); pace from an async generator instead.
+`AsyncStreamingClient` mirrors `StreamingClient` with async methods. It's safe to use as an async context manager — `disconnect()` runs on block exit even if user code raises. Pace the audio from an async generator so the event loop is never blocked.
 
 ```python
 import asyncio

assemblyai/__init__.py

@@ -1,4 +1,3 @@
-from . import extras
 from .__version__ import __version__
 from .async_client import AsyncClient
 from .client import Client
@@ -149,8 +148,6 @@
     "WordSearchMatch",
     # package globals
     "settings",
-    # packages
-    "extras",
     # version
     "__version__",
 ]

setup.py

@@ -27,9 +27,6 @@ def get_version() -> str:
         "typing-extensions>=3.7",
         "websockets>=11.0",
     ],
-    extras_require={
-        "extras": ["pyaudio>=0.2.13"],
-    },
     classifiers=[
         "Development Status :: 3 - Alpha",
         "Intended Audience :: Developers",