Meeting Assistant listens to your microphone and to the audio of the call at the same time, writes both as a labelled transcript, and does the recognition on the Mac itself.
Direct download · version 0.1.0 · 3.6 MB · macOS 26 or later, Apple silicon · notarized by Apple.
How it helps
It records the audio your Mac plays with a Core Audio tap and your microphone separately, so lines are labelled Me and Them. Nothing joins the call and nobody sees a recording bot.
Apple’s built-in speech engine starts instantly. Whisper models run on the Neural Engine or the GPU and can be compared on a built-in Benchmark page with your own recording.
The transcript scrolls as people talk, a scratchpad keeps the key points, and every meeting is saved on your Mac, with a short title, ready to export as Markdown.
Yes. It captures the audio on your Mac, so there is nothing to invite to the call.
No. Speech recognition runs on the Mac. Audio is never uploaded; a Whisper model is downloaded once, then works offline.
Italian, English, French, German, Spanish, Portuguese and Dutch, or automatic detection.
Yes, macOS 26 or later on a Mac with Apple silicon.
