Teach my Mac someone's voice or face
Teach this person's Mac to recognise someone, by a recording of their voice, a picture of their face, or a voice it already heard while listening, so what they say is named as theirs
Teach this person's Mac to recognise someone, by a recording of their voice, a picture of their face, or a voice it already heard while listening, so what they say is named as theirs
Runs on the person's Mac. Give name, kind (voice or face) and consent (who agreed to be known here, and how: nobody is enrolled without it), and one of: path (a file on the Mac), file (a file in their Trillion), or heard_voice (a voice the Mac heard while listening, like voice-2, from the listening that was running at heard_at, the time a turn of that voice was said, else the latest unless session names one); everything that voice already said in that listening is then named as theirs too. me true marks the person themselves, so their words arrive as theirs. A face enrollment keeps their picture. Several samples make it surer: enroll again to add one.
Connect the AI you already use to Trillion once. Then ask in your own words; your AI finds this tool by what it does and runs it.
Say to your AI: “Use Trillion to teach this person's Mac to recognise someone, by a recording of their voice, a picture of their face, or a voice it already heard while listening, so what they say is named as theirs”
What your AI calls: run_tool {"tool_id":"tool:enroll-someone-on-this-mac"}