Say it in pieces. Get back one piece of writing.
Most apps take one recording and hand back one transcript. unmuss keeps the thread: say a piece now, add another on the walk home, and every take is woven into the ones before it — until scattered fragments become one piece of writing that actually holds together.
Free to start · iOS & watchOS · No account required
Try it right now — no signup
Record a few seconds or drag in an audio file. unmuss transcribes and restructures it on the spot.
The demo uploads your audio to our servers, where third-party AI service providers transcribe and rewrite it on our behalf.
Preview mode — the backend isn't connected yet, so this demo returns sample text.
That's all 3 rounds of the web trial — in the App, your idea keeps growing: capture anytime, and every take weaves into the ones before it.
Download on the App StoreThe web demo is a trial with limits. The full workflow — Idea Sessions, Master Recap, Apple Watch capture — lives in the app.
How an idea grows
-
1. Say one piece. Talk for ten seconds or a minute. You get a clean, structured take — not a word-for-word transcript. (Each take is up to 60 seconds: one thought at a time, by design.)
-
2. Add another piece whenever it shows up. Record again in the same Idea Session. Your new take fuses with everything you said before, so the draft gets fuller instead of starting over.
-
3. Let the fragments pile up. A thought from your Watch, one sentence on the Action Button, a memo caught while driving — they land as Pending Materials, waiting to be woven into your next take.
-
4. Pull it all together. Master Recap turns the entire thread — every take, every fragment — into one finished draft. No new recording required.
More than transcription
-
AI that restructures, not transcribes — reads what you meant, rebuilds the logic, turns spoken language into written language. Your voice and intent stay intact.
-
Master Recap — one finished draft from a whole Idea Session.
-
Version history with side-by-side diff — every pass saved, changes highlighted, step back to any version. Nothing overwritten, nothing deleted.
-
Capture from anywhere, into the same thread — Apple Watch, Action Button, Share Extension (audio from Voice Memos), hands-free driving via Siri.
-
Context you didn't have to type — optional location and weather on each take, so a voice journal reads like a memory.
-
Eight languages — English, German, French, Spanish, Portuguese, Japanese, Chinese, Cantonese.
-
Built for your Apple kit — Live Activities, Dynamic Island, widgets, iCloud sync. Live transcription runs on-device.
-
Privacy-first — no account, no sign-up. AI rewriting is processed on our servers, where third-party AI service providers transcribe and rewrite the audio on our behalf — and you're asked for permission before your first recording is uploaded.
Built for people who think out loud
- Writers, founders, researchers — growing one idea over days.
- Commuters — best thinking happens behind the wheel; capture it hands-free.
- Professionals — turn meeting ramblings into usable notes.
- Anyone whose first draft is a ramble — and knows it.
Why not just use a transcription app?
| Raw transcript | unmuss |
|---|---|
| "Um, so I was thinking, like, maybe we should, you know, launch in, uh, March?" | "We should launch in March." |
| Take 1: "budget… Sarah… follow up?" / Take 2 (next day): "oh and the timeline slipped" | One fused note: "Discussed the budget. Sarah will follow up. Note: timeline slipped." |
Frequently asked questions
- Is unmuss free?
- Free to download. Extra credits and Unmuss Pro are available as in-app purchases. See the App Store page for current details.
- What happened to VoiceRefine?
- VoiceRefine is now unmuss. Same app, new name. (This note stays until the public App Store listing completes the rename.)
- Do I need an account?
- No. Open it and talk. No sign-up, no login, no verification codes.
- Which devices are supported?
- iPhone and Apple Watch, running iOS 18.5 or later. unmuss is Apple-only — there's no Android version.
- What languages are supported?
- Eight output languages: English, German, French, Spanish, Portuguese, Japanese, Chinese, and Cantonese. Speak in yours; the AI writes in it.
- How is my voice data handled?
- The web demo uploads your recording to our servers, where third-party AI service providers transcribe and rewrite it on our behalf. You're asked for permission before your first upload. On-device live transcription is an app-only feature. Full policy: /privacy.
- What is an Idea Session?
- A thread where short spoken takes (up to 60 seconds each) fuse into one growing draft. Say a piece now, add another tomorrow — each new take weaves into the ones before it instead of starting over.
Finally be understood.
Download unmuss and turn your next voice note into writing worth keeping.
Download on the App Store