Repository navigation
v0.1.0 beta
Pre-release
Pre-release
First public beta of Nakamura Voice — dictation (STT) and read-aloud (TTS) for OpenCode.
What's included
- Dictation (STT) — microphone in the composer; OpenRouter, Groq, or any OpenAI-compatible local Whisper server.
- Read aloud (TTS) — free Microsoft Edge voices with zero configuration, streaming playback, pause/resume.
- Auto read-aloud — reads each finished reply when the session goes idle.
- Optional local Kitsune-TTS — five anime voices, offline, with a one-click pack in
tools/kitsune-local/. - Docs in Portuguese and English — quick start for non-technical users.
Install
opencode plugin add github:NakamuraIA/opencode-plugin-voiceOnce the npm release is available:
opencode plugin add opencode-plugin-voiceThen open Settings → Voice, pick a voice and press Play sample.
Notes
- Read-aloud needs no API key (Edge TTS).
- Dictation needs an OpenRouter/Groq key or a local Whisper server.
- This is a beta release; the GUI integration uses OpenCode's experimental GUI API.
Licensed MIT.