Voicevox FAQs

Question 1

How do I install and configure Voicevox MCP Server?

Accepted Answer

Clone the repository, install dependencies with `npm install`, configure environment variables in the `.env` file (like `VOICEVOX_API_URL` and `VOICEVOX_SPEAKER_ID`), and run the server with `npm run build` and `npm start` (Windows) or via Docker Compose.

Question 2

What should I do if I encounter a connection error with VOICEVOX?

Accepted Answer

Ensure that the VOICEVOX ENGINE is running and that the API URL in your `.env` file or `mcp.json` configuration is correct. Restart the server if necessary.

Question 3

What is Voicevox MCP Server?

Accepted Answer

Voicevox MCP Server allows you to use VOICEVOX-compatible speech synthesis engines (VOICEVOX, AivisSpeech, COEIROINK) via the Model Context Protocol (MCP). It enables integration with tools like Cursor for agent-based voice synthesis.

Question 4

What are the system requirements for Voicevox MCP Server?

Accepted Answer

For Windows, you need Node.js 18+, VOICEVOX ENGINE, and VLC Media Player. For Docker, you need Docker, WSL2, VOICEVOX ENGINE, and necessary Linux libraries for audio support.

Question 5

How do I find the correct speaker ID for my voice model?

Accepted Answer

The speaker ID depends on the VOICEVOX model you're using. You can find a list of speaker IDs by accessing the `/speakers` endpoint of your VOICEVOX ENGINE API (e.g., `curl http://localhost:50021/speakers`).

Voicevox

Voicevox

主要功能

使用案例

主要功能

使用案例