Transcription & text-to-speech
Auto-transcription (STT)
Every audio message gets a text transcript so you can scan a room without listening end-to-end. Powered by OpenAI Whisper or Groq depending on configuration.
Text-to-speech (TTS)
Text messages can be auto-spoken in the room. Toggle in Settings → Profile (“Generate TTS for my text messages”).
Bring-your-own keys
In Settings → Voice, you can plug in your own OpenAI / Groq API key for transcription and TTS — useful for power users who want a specific model or higher rate limits.
Per-room overrides
Room admins can pin a specific STT provider for compliance-sensitive rooms (e.g. lock to a self-hosted backend regardless of what individual users configured).