Dettato records Teams, Zoom, Webex, and browser calls, then creates transcripts, speaker-aware summaries, and action items on your Mac. Start free with file transcription and dictation.
No account or card · macOS 26+ · Apple Silicon · Pro is $39 USD once with a 14-day refund
Try it Click a recording, run Summary, hit Stop on the live call — the real interface, rebuilt for the web.
The whole point
Recording, transcription, speaker identity, and AI summaries all run on your Mac's own silicon. No Dettato server ever sees your meetings — there isn't one to send them to.
Your mic and the meeting app's audio, tapped straight from Core Audio. Survives crashes and AirPods swaps.
Whisper and Parakeet on the Neural Engine, a local LLM on Metal — all of it on your Mac's own silicon.
Your recordings and library live in one folder on your disk. Search it, export it, delete it. Yours.
Audio uploaded to our cloud: some none. No accounts, no analytics, no telemetry. Out of the box, the app's network activity is model downloads, a license check, and — only with your OK — update checks. Cloud engines, webhooks, and URL imports run only if you set them up. The privacy policy is short enough to actually read.
Quietly capable
Built for people who live in calls but keep what's said to themselves.
Detects calls in Teams, Zoom, Webex, and browser meetings — catches the start, stops when you hang up. Or press record yourself; auto is optional.
Apple Speech, Whisper, or Nvidia Parakeet on Apple Silicon. Cloud engines only if you bring your own keys.
On-device diarization separates voices; voiceprints suggest recurring people. It only suggests — you stay in charge of names.
A local LLM writes the recap and pulls action items with owners. Or point it at your own OpenAI-compatible endpoint — your key, your provider, your call.
Yank your AirPods and the recording rides through it. Crash mid-call, and Dettato recovers everything on relaunch — worst case costs minutes, never the meeting.
Drop in a .vtt or .srt caption file from Teams, Zoom, or Meet — speakers and timings intact, no audio needed. Audio files transcribe on import too.
Watch folders transcribe the audio you drop in them; the dettato CLI scripts the rest.
Young, and honest about it
What ships today is the core we trust: recording, transcription, speakers, summaries. Around it, Dettato improves in small, steady releases.
Auto-update is built in. New versions arrive in the app — no re-downloading, no hunting for installers.
Every 1.x update — features and fixes alike — is included with Pro. No upgrade fees, no subscription meter.
Each release says plainly what changed and what got fixed. No “bug fixes and improvements” hand-waving. Read it →
Dettato is independent software from a very small team. That means fast, focused releases — and it also means support is by email and best-effort, not a call center. We think that's a fair trade, and we'd rather say so up front.
Try the workflow first
Import a recording or use dictation without an account or card. Pro adds automatic meeting capture, speaker identity, AI assist, and watch folders.
Prefer drag-install? Get the .dmg
Polar shows your local equivalent and applicable tax before payment. Your key arrives by email. 14-day refunds, no forms.
The .pkg installer also ships the dettato command-line tool — it stays free and ungated.
Questions or payment trouble? support@dettato.app — a person reads it.
Fair questions
No account, no card, and no meeting bot. Upgrade only if automatic capture is worth it.