Hear what a device recorded
Hear a piece of sound another device recorded while listening (a phone, a watch, a browser, glasses) and send every turn in it to Trillion as its speaker's words, named against the people this person's Mac knows
Runs on the person's Mac with the same hearing, the same enrolled voices and the same classifier as the Mac's own listening. Give file (the piece, kept in their Trillion) or path (on the Mac), at (when the piece started) and device (the device that recorded it, like iPhone). Each turn goes to the one door as Listening, in that device's own conversation; a voice enrolled on any device is named here, and an unnamed voice keeps its number across that device's pieces. Answers { sent, left_out, waiting, naming, error }.
What it asks for
- file (Optional, words): The recorded piece in the person's Trillion, by name or identity
- path (Optional, words): The recorded piece on the Mac
- at (Optional, words): When the piece started recording
- device (Needed, words): The device that recorded it, like iPhone or Apple Watch
Use it from your AI
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 hear a piece of sound another device recorded while listening (a phone, a watch, a browser, glasses) and send every turn in it to Trillion as its speaker's words, named against the people this person's Mac knows”
What your AI calls: run_tool {"tool_id":"tool:hear-for-a-device"}