Real-time voice-to-text, with one auto-applied AI Action.
Hold the hotkey, talk, text appears at your cursor in any app on macOS — already polished by one Action you picked (Polish, Translate, or your own). Local-first, no language picker, no cloud round-trip.
You're mid-sentence in Mail, Slack, or your editor. You hit the hotkey, you keep thinking out loud, and the text lands where your cursor is. The flow doesn't break.
Why streaming feels different.
Local-first, by default
Streaming uses on-device transcription on Apple Silicon — audio never leaves your Mac. Zero cloud round-trip, zero latency tax, fine on a flaky hotel Wi-Fi.
Anywhere your cursor is
Text inserts at the macOS input focus — Mail, Slack, VS Code, Notes, a Figma comment. No browser extension, no per-app integration. If you can type there, Quilin can dictate there.
Built for thinking speed
Streaming keeps up with mid-sentence pauses, accented English, and mid-thought correction. Talk at your natural pace — Quilin doesn't make you slow down to be understood.
Roles who lean on this most.
Two moments from the full day-in-the-life. The whole roster lives at /scenarios.
- Engineer / Designer / PMWed · 9:42 AM
Write a PR description faster than you can type
You just pushed the migration. You hit the hotkey at your desk, talk through what changed and why, and the description writes itself in EN. Reviewers in Shanghai get the same context — clean transcript, code-switching kept intact.
See Engineer / Designer / PM day → - Founder / OperatorMon · 6:15 PM
End-of-day broadcast in two languages
After the last 1:1, you record a five-minute voice update at your desk. Run "Translate to ZH for the Shanghai team" → both versions ship before you close the laptop.
See Founder / Operator day →
Related capabilities
Get streaming dictation on your Mac.
Quilin opens beta in June 2026. Join the waitlist for an early invite.