Voice
Voiceclaw
OpenClaw voice assistant plugin β wake word, STT, TTS
Install
openclaw plugins install @openclaw/voiceclaw
Configuration Example
plugins:
entries:
voiceclaw:
enabled: true
config:
stt:
engine: sensevoice
language: ko
tts:
engine: edge-tts
voice: ko-KR-SunHiNeural
wakeWord: "hey claw"
autoStart: false
README
# ποΈ VoiceClaw
**OpenClaw voice assistant plugin** β wake word detection, speech-to-text, text-to-speech, all as a standard OpenClaw plugin.
## Features
- π£οΈ **Wake word** β "Hey Claw" triggers voice interaction
- π€ **STT** β SenseVoice (primary), faster-whisper, OpenAI Whisper
- π **TTS** β Edge TTS, Piper, macOS `say`
- π§ **Agent integration** β Voice input goes directly to your OpenClaw agent
- βοΈ **Standard plugin** β `openclaw.plugin.json` manifest, config via `openclaw.yaml`
## Architecture
```
Node.js (OpenClaw plugin) βββ JSON-RPC stdio βββΊ Python (audio/ML)
βββ service.ts (lifecycle) βββ vad.py (Silero VAD)
βββ bridge.ts (IPC) βββ stt.py (SenseVoice/Whisper)
βββ commands.ts (/voiceclaw) βββ tts.py (Edge TTS/Piper)
βββ gateway-methods.ts (RPC) βββ audio.py (PyAudio)
βββ tools.ts (voice_speak) βββ server.py (JSON-RPC server)
```
## Quick Start
```bash
# Install
openclaw plugins install @openclaw/voiceclaw
# Setup Python deps
openclaw voiceclaw install
# Start
openclaw voiceclaw start
# or in chat: /voiceclaw start
```
## Config
In `openclaw.yaml`:
```yaml
plugins:
entries:
voiceclaw:
enabled: true
config:
stt:
engine: sensevoice
language: ko
tts:
engine: edge-tts
voice: ko-KR-SunHiNeural
wakeWord: "hey claw"
autoStart: false
```
## Design
See [DESIGN.md](DESIGN.md) for full architecture, bridge protocol, and component details.
## Status
π§ **Design phase** β migrating from [openclaw-voice](https://github.com/muin-company/openclaw-voice) to plugin standard.
## License
MIT
voice
Comments
Sign in to leave a comment