Uses your phone’s speed to detect driving and switch to a fully hands-free voice view: it listens continuously, shows a live transcript as you speak, sends automatically when you pause, and reads replies aloud — no tapping. Voice commands work too: say “stop”, “new chat”, “never mind”, or “exit hands-free”. Prefer it to respond only when addressed? Say “go to sleep” (or tap “Hey Goose” mode) and it ignores everything until you say “Hey Goose” — handy with passengers; “open mic” switches back. You can also talk over a spoken reply: “stop” or “Hey Goose…” interrupts it. Live transcription uses whichever speech engine you pick under Speech recognition below; location always stays on your device. The screen stays awake while hands-free; works while this tab is open (not in the background).

Which engine turns your speech into text. Auto uses accurate cloud transcription (Deepgram, tuned to your project’s vocabulary) when your box has it, and your device’s built-in recognizer otherwise. Cloud forces Deepgram; On-device keeps audio on your device (no speech sent to us) using the browser recognizer.

End-of-turn sensitivity — how soon hands-free decides you’ve finished talking and sends (cloud engine only).