devwithjev
— reading now— views
Submit a build

Voice-Controlled Browser Automation

Jev (TypeSafe System One) decides intent and target from voice input, while Playwright acts on a real browser. The project says this takes ~300 ms per spoken word, often before the user finishes speaking.

Open source time~300 ms per spoken word
moritzkremb@moritzkremb
Control a real browser by voice. Jev (TypeSafe System One) decides intent + target in ~300 ms per spoken word; Playwright acts — often before you finish the sentence.
Sep 19, 2026GitHubView on GitHub

Also filed under Agents & browsers