Meeting Transcription
The best meeting transcription is the one nobody notices. No bot participant. No notification. No awkward “is that recording?” moments. Just a complete transcript when the call ends.
The problem
Otter, Fireflies, and every other cloud service work the same way: a bot participant joins your meeting. Everyone in the room sees it. A notification announces it.
On internal standups, that's fine. On a client call, an investor pitch, a board meeting, a personnel discussion — it changes the room. People speak differently when they know they're being recorded by a third party.
And every word of that conversation travels to someone else's servers, processed on infrastructure you don't control, retained under policies you didn't write.
What everyone sees
Everyone in the meeting sees the bot. Everyone knows.
With Resonant
No bot. No notification. Resonant records from your Mac's audio — invisible to participants.
How it works
Mic captures your voice. System audio captures everyone else. Two channels, transcribed separately, merged with speaker labels. No bot participant added to the call.
NVIDIA Sortformer identifies speakers from the system audio channel. Combined with the mic/system split, you get accurate attribution without sending a single byte to the cloud.
The recording indicator hides automatically when you share your screen. No one sees a badge. No one asks questions. The transcription continues silently.
Resonant knows when you join a call in Zoom, Meet, Teams, or any supported app. Recording starts automatically. No setup, no remembering to hit record.
Works with
Resonant captures system audio. If you can hear it, Resonant can transcribe it.
Comparison
After the call
Full-text search across every meeting. Find what Sarah said about the deadline three weeks ago. Speaker labels, timestamps, topic headings — all indexed locally.
Any MCP-compatible agent — Claude, Codex, and others — can pull meeting context directly. Ask your AI assistant what the team decided and get the exact quote.
Meeting transcripts flow into your Resonant workspace alongside dictation, notes, and voice memos. One searchable record of everything you said and heard.
Privacy
Cloud meeting tools send your audio to their servers for processing. Every word of every meeting — client negotiations, board discussions, HR conversations — passes through infrastructure you don't control.
Resonant runs transcription and diarization models directly on your Mac. Audio is processed in memory and never written to disk unencrypted. The transcript stays on your device. No vendor sees it. No vendor stores it. No vendor trains on it.
You decide what to keep, what to search, what to export, and what to delete. That's the architecture. There's no setting to toggle. There's no server to trust.
Free. Local. No bot required.
Dual-channel diarization, speaker labels, searchable transcripts. All on your Mac. No subscription, no cloud, no bot.
Requires macOS 14+ · Apple Silicon