OpenQuack Speak. Send. Privately.
Voice dictation for macOS. Nothing leaves your device — audio, text, nothing.
OpenQuack is a tiny menu-bar app for macOS. Press a hotkey, speak, press it again — your transcript appears at the cursor. Wherever you can type, you can talk.
Speech recognition happens on your Mac. No cloud, no account, no signup, no telemetry.
Local. Everything runs on your device — recording, transcription, optional polish. Nothing leaves: no audio, no text, no telemetry, no signup. Confidential work stays confidential, by construction. And because there's no API call in the loop, it just keeps working — offline, on a plane, behind a corporate firewall.
Fast. Whisper on Apple Silicon transcribes in roughly a fifth of the time you spent speaking. ~2.6% word-error rate on real human speech on a baseline M4 / 16 GB. Faster than typing in most cases. Full bench matrix in docs/BENCHMARKS.md.
Open. MIT-licensed. Every line is auditable; every change happens in public. The version running in your menu bar is the version in this repo.
- One-key dictation. Pick a hotkey (default ⌃⇧Space). Toggle or push-to-talk.
- All local. Speech recognition runs on your Mac. No internet needed for dictation — works offline, on a plane, in a tunnel, behind a corporate firewall. No API keys, no rate limits, no service outages. Same dictation in personal or business settings.
- Multi-language. Whisper handles 99 languages — English, Chinese, Japanese, Korean, Spanish, French, German, Italian, and Portuguese are right in Settings; auto-detect on by default.
- Auto-paste at the cursor in any app. (Falls back to your clipboard if you'd rather paste yourself.)
- Smart formatting — capitalisation, end-punctuation, "um/uh" cleanup.
- Custom dictionary — teach it the proper nouns and project names you actually use.
- Auto-stop after silence. Finish speaking, OpenQuack wraps up on its own.
- Live mic-level overlay so you can see it's listening.
- Quick first-launch setup — permissions, hotkey, done in a minute.
- Tiny. An 8 MB menu-bar app, plus the speech model on first run.
- Open source, MIT.
- Nothing leaves your device — audio, text, nothing. Recording and transcription are fully local. Always.
- No analytics, no telemetry, no signup.
The full privacy contract is in docs/VISION.md.
A peek at what's queued up. Both build on the dictation foundation that ships today.
In-context transcription. OpenQuack will read the surrounding text where you're about to paste — the line above the cursor, the function you're inside, the chat thread you're replying to — and feed it to the speech model as context. Domain terms get disambiguated by what you're actually doing ("cloud code" turns into "Claude Code" when you're in a terminal, not the other way around). Less custom-dictionary tinkering needed.
Thinking mode. A second pass after transcription, run through a small local LLM, that turns a raw spoken sentence into a written one you'd actually press send on. Filler trimmed, structure tightened, the right capitalisation on words that matter. Off by default, one-toggle opt-in. Fully local — Ollama or MLX-LM, your pick.
Schedule and SPEC details in docs/ROADMAP.md.
The duck has bigger plans. Where this is going: docs/VISION.md.
brew tap larryxiao/openquack https://github.com/larryxiao/openquack
brew install --cask openquackOr download the DMG and drag into Applications. First launch: right-click → Open → Open (one-time Gatekeeper bypass).
Grant Microphone when macOS asks, pick a hotkey in Settings → Shortcut (default ⌃⇧Space).
Want a guided walkthrough? See docs/TUTORIAL.md — five minutes from install to first dictation.
Paste this into Claude Code, Codex, opencode, Hermes, or similar:
Install OpenQuack on this Mac:
brew tap larryxiao/openquack https://github.com/larryxiao/openquack
brew install --cask openquack
(Or grab the DMG from https://github.com/larryxiao/openquack/releases
and drag it into /Applications; first open right-click → Open → Open.)
Then launch /Applications/OpenQuack.app, grant Microphone, and pick a
hotkey in Settings → Shortcut. Default ⌃⇧Space.
More options (uninstall, build-from-source, what's downloaded on first run): docs/INSTALL.md.
OpenQuack is AI-native open source — every PR cites a SPEC, atomic tasks come from the roadmap, the workflow is friendly to coding agents at scale (and humans on the same path).
Start with AGENTS.md, pick a 🔵 task in docs/ROADMAP.md, open a draft PR.
Under the hood: TUTORIAL · DEVELOPMENT · ARCHITECTURE · BENCHMARKS · DESIGN · INSTALL.
MIT — see LICENSE.