Plugin · AudioSource
Shared capture with echo cancellation
One capture for both sides, with an echo-cancelled microphone when models exist.
Built on vendored capture code from the MIT-licensed anarlog project (see the repository’s NOTICE). One capture yields microphone and system audio per frame and splits into a You and a Them source.
You uses the echo-cancelled microphone when echo cancellation produced one, else the raw microphone. Without the echo-cancellation models it captures without them and reports it.
More plugins
All plugins →- 01
Device capture (BlackHole, cpal)
Captures any input device, such as the BlackHole virtual device, through cpal.
- 02
ScreenCaptureKit audio
Native macOS system audio for Them, and the microphone for You, with no driver.
- 03
Deepgram streaming
Cloud speech-to-text over one WebSocket per side, with interim results.