voice input
talk to a colony instead of typing.
- status
- built-in ships with colonizer and is on without you doing anything.
- version
- ships with the harness
- licence
- mit (part of the harness)
- docs
- configuration
what it does for you
the composer’s microphone uses the browser’s own speech recognition by default. you can connect openai, groq, deepgram, elevenlabs or any openai-style transcription server instead.
what it needs
- nothing for the browser’s recogniser (chrome sends the audio to google; safari can recognise on the device); a key for a service
how to turn it on
the microphone in the composer. change the service in settings → modules → voice.
how far it has been tested
ships in releases.