RVC-Voice-MCP
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@RVC-Voice-MCPSpeak 'Hello there' using the saved voice model"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
RVC Voice MCP
Turn text from an MCP-capable chat client into speech, convert it with a local RVC .pth + .index model, and render an audio player in the conversation.
This project generates synthetic audio. Only use voices you own or have permission to use, and clearly disclose AI-generated output.
How it works
ChatGPT calls
speak_rvcwith text.The local server asks Applio to synthesize a base voice and run RVC conversion.
The MCP Apps widget renders the generated WAV in an inline player.
The server never accepts model paths or shell commands from chat. Paths are fixed in local environment configuration and excluded from Git.
Requirements
Windows and Node.js 20+
A working Applio installation, either inside
local/Applioor at a manually configured pathAn RVC inference
.pthand matching.indexChatGPT Developer Mode and a public HTTPS tunnel for web integration
Local setup
Copy-Item .env.example .env
npm install
npm run check
npm startEdit .env before starting and replace HTTP_TOKEN with a long random secret.
Local automatic discovery
The repository reserves these ignored local directories:
local/Applio/ Complete Applio installation
local/models/ RVC .pth and .index filesIf the four path settings in .env are left empty, the server uses local/Applio/env/python.exe and searches local/models recursively. Automatic model selection is enabled only when exactly one .pth and exactly one .index file are present, preventing an accidental mismatch when several voices are installed.
Manual paths still take priority. You can set any or all of these values independently:
APPLIO_DIR=D:\Apps\Applio
PYTHON_PATH=D:\Apps\Applio\env\python.exe
RVC_MODEL_PATH=D:\Voices\character.pth
RVC_INDEX_PATH=D:\Voices\character.indexRestart the service after adding or changing models because discovery happens at startup.
Health:
http://127.0.0.1:8788/healthMCP:
http://127.0.0.1:8788/mcp?token=YOUR_TOKEN
One-click public endpoint
Run:
npm run publicThe first run downloads Cloudflare Tunnel into the ignored local tools folder. The script generates a fresh access token, starts the MCP server and tunnel in the background, then prints the HTTPS connector URL to add in ChatGPT Developer Mode. Keep that terminal open only while reading the URL; the two background processes continue running.
Stop both processes with:
npm run stopQuick tunnels are intended for development and their URL changes each time. For a stable public release, configure a named Cloudflare Tunnel and a domain you control.
Tools
speak_rvc(text)generates AI speech and renders the player.get_status()checks local configuration without exposing paths.
Current scope
The initial adapter uses Applio's built-in TTS command and RVC inference. Fully offline TTS and a stable named tunnel are future work.
License
MIT
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Connectors
MCP server exposing the AceDataCloud Fish Audio API (text-to-speech with voice conditioning)
MCP server for AI dialogue using various LLM models via AceDataCloud
MCP server for Producer/Riffusion AI music generation
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/QingMo-A/RVC-Voice-MCP'
If you have feedback or need assistance with the MCP directory API, please join our Discord server