Your voice becomes clean, finished text in any app on your device — in any language, even two languages mixed in one sentence. Fillers removed, punctuation placed, your words kept.
Say the sentence the way you'd say it — Semoru keeps each word in its own script. Flip one setting and it translates everything into whatever language you want, as you speak.
Casual in Slack, proper in Gmail, untouched in your code editor. Semoru learns the surface of how you write — greetings, emoji, sign-offs — and never rewrites what you meant.
hi Rahim new line the first cut is ready new paragraph we deliver Thursday scratch that we deliver Monday
courier please, deliver to my address before noon
Addresses, meeting links, sign-offs — anything you repeat but never want to spell out loud. Works in any language you dictate in.
A dictation tool hears your work, your names, your numbers. That demands more than a policy page.
On Apple silicon, speech becomes text on your Mac. Airplane mode? Still works.
When the server is used, audio streams into memory, is transcribed, and the buffer is freed. No disk, no logs. Transcripts live only on your machine.
Open models on our own servers. Your words are never routed to an AI company's API.
A live log in the app shows every request: what was sent, how large, which region. It's usually empty.
Semoru cannot see your screen. Not a setting. Not a toggle. Absent.
No card. No trial clock. Beta testers dictate free.
Requires macOS 12+ on Apple silicon (M1 or newer). Direct download.
Drag to Applications, then right-click → Open → Open — just the first time. The app is signed, and every update is verified cryptographically.
Allow the microphone and Accessibility, sign in with your email, download the on-device model (570 MB, once).
That's the product. Updates arrive by themselves from here on.
windows & linux coming — speaking is ~3× faster than typing
Just talk — to clients, to your mom, to your code editor — and watch finished words land anywhere. In every language you live in.
Download Semoru for Mac