Local-first
Core communication works with zero network. No feature anyone relies on to speak may depend on a server, an account, or connectivity.
Speech shouldn't have a subscription
Dedicated AAC devices cost thousands; capable apps charge hundreds up front or rent by the month. Public AAC exists so that money is never the reason someone can't speak: a free, open-source communication app for people with ALS/MND, cerebral palsy, autism, aphasia, and stroke — and the families, therapists, and educators who support them.
Six principles we don't negotiate
No cloud inference
Speech, prediction — everything runs on the device. No usage-metered APIs deciding whether someone gets to talk today.
Agency over fluency
Software may propose; only the person disposes. Nothing is ever auto-spoken, and the literal selection always wins over a “smoother” rewrite.
Accessibility is the product
Touch, switch, dwell, and gaze operate the same boards. WCAG 2.1 AA is the floor, not the goal.
Your voice stays yours
Voice recordings live on your device, never train anything, and only ever leave as your own exports or end-to-end-encrypted sync you explicitly opt into.
Interoperability is mandatory
Lossless OBF/OBZ import and export from day one. Leaving must always be easy — that keeps us honest.
Free and open source, verifiably
The entire app is open source under the MIT license — every claim on this site can be checked againstthe code on GitHub. Symbol sets and voices keep their own upstream licenses. There is no company to be acquired, no pricing page waiting to appear: the project is built to stay free.
It's under active development. Bug reports, boards, translations, and code are all welcome on GitHub.