Settings

Voice input and text-to-speech

System voices, plus OpenAI / Gemini / ElevenLabs TTS. Whisper-style transcription on supported devices.

Voice features are split into two flows. Speech-to-text (you talk, Todrise types) goes through your OS's native speech recognizer where available, or an OpenAI Whisper-compatible API endpoint if you configure one. Text-to-speech (Todrise reads a note or AI reply out loud) uses your system TTS by default and can be upgraded to OpenAI, Gemini, or ElevenLabs voices in Settings → Voice.

Common setups

  • Whisper API — paste an OpenAI key, choose a model size.
  • ElevenLabs — pick any voice from your account.
  • Gemini — choose from Google's voice catalog.

Still stuck?

Open an issue on GitHub Issues.