Voicebox MCP Server
[](https://miyabi-ai.jp)
# VoiceBox MCP Server
Model Context Protocol (MCP) server for VOICEVOX text-to-speech integration.
## Features
- `voicebox_speak` - Convert text to speech
- `voicebox_status` - Check task status
- `voicebox_speakers` - List available speakers
- `voicebox_health` - Check system health
- `voicebox_metrics` - Get system metrics
## Prerequisites
1. **VoiceBox TTS API** must be running:
```bash
cd ~/dev/voicebox-tts
./scripts/start.sh
```
2. **VOICEVOX** must be running with API server enabled (port 50021)
## Installation
```bash
cd ~/dev/voicebox-mcp
npm install
npm run build
```
## Claude Code Configuration
Add to `~/.claude/settings.json`:
```json
{
"mcpServers": {
"voicebox": {
"command": "node",
"args": ["/Users/shunsukehayashi/dev/voicebox-mcp/dist/index.js"],
"env": {
"VOICEBOX_API_URL": "http://localhost:5001"
}
}
}
}
```
## Available Speakers
| ID | Name |
|----|------|
| 0 | 四国めたん (あまあま) |
| 1 | 四国めたん (ノーマル) |
| 2 | 四国めたん (セクシー) |
| 3 | ずんだもん (ノーマル) |
| 4 | ずんだもん (あまあま) |
| 5 | ずんだもん (悲嘆) |
| 6 | 春日部つむぎ (ノーマル) |
| 7 | 春日部つむぎ (あまあま) |
| 8 | 春日部つむぎ (ツンデレ) |
## Usage in Claude Code
```
Please use voicebox_speak to say "Hello, world!" with speaker 3 (Zundamon).
```
```
Check the voicebox system health.
```
```
List all available VOICEVOX speakers.
```
## Development
```bash
npm run dev # Watch mode
npm run build # Build
npm start # Run server
```
TDQS
Scored across 5 tools
Each tool has a clearly distinct purpose: health checks system status, metrics provides statistics, speak converts text to speech, speakers lists available voices, and status checks task progress. There is no overlap in functionality, making tool selection straightforward for an agent.
All tools follow a consistent 'voicebox_' prefix with descriptive suffixes (health, metrics, speak, speakers, status). This uniform naming pattern enhances predictability and readability across the toolset.
With 5 tools, the server is well-scoped for a TTS system, covering essential operations like health checks, metrics, speech generation, speaker management, and task monitoring. Each tool serves a clear, necessary function without redundancy.
The toolset covers core TTS workflows effectively, including system monitoring, speech synthesis, and task management. A minor gap exists in lacking tools for advanced operations like canceling tasks or managing speaker configurations, but agents can work around this with the provided tools.