Skip to main content
Octo supports voice interaction via ElevenLabs TTS and Whisper STT.

​
Setup

ELEVENLABS_API_KEY=...
ELEVENLABS_VOICE_ID=...   # optional, uses default voice

​
Usage

Toggle voice mode: 
/voice on     # enable TTS for responses
/voice off    # disable TTS
Or start with voice enabled: 
octo --voice

​
Telegram Voice

When using the Telegram transport:
  • Incoming voice messages are transcribed via Whisper
  • Responses are sent back as voice messages when voice mode is enabled
  • Text responses are always sent alongside voice for accessibility