Features
- Local processing — recording, transcription and summaries all run on your Mac. Your audio is never uploaded anywhere.
- Call detection — spots a Zoom, Google Meet (Chrome) or Slack huddle and offers to start recording.
- AI summaries — key points, action items, decisions extracted automatically.
- Full transcripts — searchable, exportable text of every meeting.
- Who said what — your mic and the call audio are recorded as two streams, so every line is marked as you or as the other side.
- Menu bar app — one click to start recording, stays out of your way.
FAQ
Does it send my audio anywhere?
No. Recording, transcription (Whisper) and summarization (a local Ollama model) all happen on your Mac, and no audio or transcript is ever uploaded.
Three things do reach the network, none of them your content: the one-time model download on first run, a check for app updates, and crash reports — which are off unless you turn them on.
Do the summaries need anything installed?
They need Ollama, which is free and runs locally. First-run setup installs it for you if you have Homebrew; otherwise you install it yourself. Without it you still get the full transcript, just no summary.
What Macs are supported?
Requires macOS 13 (Ventura) or later.
Can it record system audio?
Yes. It captures both microphone and system audio, so it works with any meeting app (Zoom, Google Meet, Teams, etc.).
How accurate is the transcription?
It records with Whisper's base model for the live view, then re-transcribes the whole recording with the larger medium model when you stop. Clean audio gives the best results.
