packages/kilo-docs/pages/code-with-ai/features/speech-to-text.md
{% callout type="warning" title="Experimental feature" %} Speech to Text is experimental. Expect issues and changes as it matures. {% /callout %}
Use voice input in prompt fields instead of typing. Transcription uses your Kilo account through Kilo Gateway.
Voice input needs FFmpeg plus access to the Kilo provider.
FFmpeg is required for audio capture and processing. Install it for your platform:
macOS:
brew install ffmpeg
Linux (Ubuntu/Debian):
sudo apt update
sudo apt install ffmpeg
Windows: Download from ffmpeg.org/download.html and add to your system PATH.
Enable and sign in to the Kilo provider to use voice input in prompt fields. Requests use your Kilo account through Kilo Gateway, so no separate OpenAI provider profile or API key is needed.
Voice input is experimental and must be enabled:
Kilo stores this toggle in your global Kilo CLI config (~/.config/kilo/kilo.jsonc), not VS Code user settings:
{
"experimental": {
"speech_to_text": true
}
}
Once enabled, a microphone button appears in prompt fields:
The feature includes real-time audio level visualization and voice activity detection to automatically detect when you're speaking.
Microphone button not appearing:
Transcription errors:
Speech to Text is experimental and may have limitations: