talk-to-me
Speak the answer instead of printing it, when reading the screen is the problem.
Inputs
- text — What to say. Rewritten for the ear before synthesis, not read verbatim.
- voice — Neural voice name, e.g. en-GB-SoniaNeural. Defaults to the British female voice.
- content — The material to brief on — a task list, a decision set, a status summary
- batches — How many batches to split into. Defaults to one per natural grouping, 3-5 items each.
- trigger — No input needed; run when audio fails or on a new machine
- state — Whether to start or stop speaking answers by default (True | False)
How you ask
- Talk to me — I can't read my screen right now
- Read that summary out loud
- Give me the briefing in audio, in batches so I can stop you
- Stop speaking, back to text
Outputs
- console
- mode-acknowledgement (text)
- played in sequence, one Bash call per batch
- played through the platform's audio player; the mp3 is deleted after playback
- setup-report (text)
- spoken-audio (audio)
- spoken-briefing (audio)
Levelbeginner
Inputconversation
Outputdocument
Setup[none]
Roles[CEO / Founder]