A voice when you want one.
Talk, record a thought, or explicitly enable local hands-free detection. The modes have different limits.
Tap to talk
Documented · local Tap the mic or press Alt+M and talk it through. Local recording stops on silence, another tap, or after 15 seconds. Tab moves keyboard focus. Optional wake detection has its own setting and starts off. After recording, “Got it” means the recogniser was confident, “I may have misheard” means it was not, and “Check the words” means confidence was unavailable. Review the words before Send.
Notes & ideas
Documented · local Catch an idea before it wanders off. Type a note or record up to 5 minutes, review the transcript, then choose Clean up, Develop idea, or Project checkpoint. Originals stay separate from AI drafts. Processing stays local in every mode; long notes can take several minutes.
Use Stop & transcribe or Discard clip. After speech starts, thinking pauses do not end a Notes clip; a clip with no initial speech ends after ten seconds. Add another clip after transcription. Saved text and revisions are included in backups and exports; deleting a note does not remove earlier copies.
Hello Phantom · optional
Optional · owner enabled Enable local “Hello Phantom” detection in Settings, say the phrase, wait for Listening, then ask a general question. Off by default and English-only. General wake questions have no personal memory, documents or action tools. Starting cooking on the unlocked screen can separately authorize Kitchen’s fixed hands-free controls for that session. Workshop can separately authorize hands-free searches of explicitly checked manuals and fixed procedure controls. Kitchen and Workshop use different session modes; locking, expiry, restarting or a Workshop source change ends that access. Job records have no hands-free reading permission. Noise can cause misses or false triggers; room reliability has not been established.
Detection pauses during device speech, manual recording, setup and maintenance. Turning it off stops passive capture. This is a software setting; unplugging the USB microphone disconnects the input physically. A detected phrase does not identify the speaker.
Speech and language
Documented · local Setup and Settings select the preferred language for written answers. Speech recognition and spoken replies work in English, Italian and Spanish (Italian and Spanish were tested with synthetic voices; real speakers vary); the other written languages are text-only. Replies follow the language you speak or type by default; Settings → Reply in can keep them in the selected language. English offers four voices; Italian and Spanish each offer one, with three speaking speeds and a preview in all three languages.
Documented · local In Notes, Read aloud speaks the note on the device; a ready document is read page by page in the voice of the language it is written in (English, Italian or Spanish). Tap the same button again to stop. With voice off nothing is read. Check its status.