Skip to main content
Glama
OpenUtau-MCP

OpenUtau-MCP

by turboegg1145

OpenUtau-MCP ๐ŸŽค๐Ÿค–

Standalone Model Context Protocol (MCP) Server for OpenUtau & DiffSinger
Enables AI Agents (Claude Desktop, Cursor, Antigravity, VS Code) to directly generate, inspect, and micro-tune OpenUtau v0.6+ (.ustx) project files and DiffSinger neural expressions.


๐ŸŒŸ Features

  • ๐Ÿ› ๏ธ Native .ustx Project Generator: Builds multi-track YAML project files with Singer IDs and Phoneticizer bindings.

  • ๐ŸŽ›๏ธ DiffSinger & Neural Expression Curves: Generates Dynamics (dyn), Tension (tns), Breathiness (bre), and Vibrato curves.

  • ๐Ÿ” Project Inspector: Parses existing .ustx files, extracting track structures, notes, and metadata.

  • ๐ŸŽถ Acoustic Preview Synthesizer: Instant 24-bit PCM preview audio synthesis with 25ms cosine-squared crossfading.

  • โšก Zero-Dependency Core: Lightweight Python stdio JSON-RPC 2.0 server.


Related MCP server: utauMCP

๐Ÿš€ Quick Setup

1. Install Dependencies

pip install -r requirements.txt

2. Verify Server

python test_mcp.py

3. Add to your AI Agent Config (mcpSettings.json)

{
  "mcpServers": {
    "openutau-mcp": {
      "command": "python",
      "args": ["F:/antigravity lol/github้กน็›ฎ/OpenUtau-MCP/server.py"]
    }
  }
}

๐Ÿ“„ License

MIT License (c) 2026 turboegg1145

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelogโ€ฆ

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      D
      maintenance
      Exposes 16 music21 MIDI analysis and editing tools to AI agents, enabling them to read, edit, generate, and analyze MIDI files through natural language.
      MIT
    • F
      license
      Not graded
      quality
      D
      maintenance
      Controls OpenUtau (vocal synthesis software) from Claude Desktop, enabling project creation, editing, and live note manipulation via a bridge plugin.
      3
      -
    • A
      license
      B
      quality
      D
      maintenance
      Enables coding agents to compose, tune, render, mix, and audit native VOCALOID3/4 projects from scratch, acting as a production bridge between intent and finished song.
      22
      14
      MIT
    • A
      license
      Not graded
      quality
      C
      maintenance
      Enables AI agents to create and edit Strudel music code, render offline WAV audio, and obtain structured hearing reports with waveform, spectrogram, BPM, and onset analysis.
      AGPL 3.0