MCP Audio Server

MCP Audio Server

A Model Context Protocol (MCP) server for audio processing and chord analysis.

Quick Start

Use Docker Compose for the quickest setup:

# Clone the repository
git clone https://github.com/yourusername/mcp-audio-server.git
cd mcp-audio-server

# Start the server
docker compose up -d

Sample Request

# Analyze a WAV file
curl -X POST http://localhost:8000/analyze_chords \
  -H "Content-Type: application/json" \
  -d '{
    "audio_data": "'$(base64 -i tests/fixtures/tempo/120bpm_with_chord.wav)'",
    "format": "wav",
    "options": {"model": "basic"}
  }'

Sample response:

{
  "schema_version": "1.0.0",
  "key": "C",
  "tempo": 120.5,
  "chords": [
    {
      "time": 0.0,
      "label": "C",
      "confidence": 0.92
    },
    {
      "time": 1.0,
      "label": "G",
      "confidence": 0.87
    }
  ],
  "duration": 4.0,
  "processing_info": {
    "sample_rate": 44100,
    "channels": 1,
    "processing_time": 0.245,
    "model_used": "basic"
  },
  "correlation_id": "550e8400-e29b-41d4-a716-446655440000"
}

Features

• Audio file decoding and normalization with FFmpeg
• Music analysis capabilities:
  • Tempo detection (BPM)
  • Key detection
  • Chord analysis and tracking
• RESTful API with structured responses
• JSON schema validation for inputs and outputs
• Robust error handling with descriptive messages
• Resource management with concurrency controls
• Caching for performance optimization
• Observability with structured logging and metrics
• Comprehensive test suite
• Containerized deployment with Docker

Installation

Docker (Recommended)

docker compose up -d

Manual Installation

1. Prerequisites:

   • Python 3.10 or higher
   • FFmpeg
   • Poetry

2. Install dependencies:

   poetry install
   

3. Run the server:

   poetry run uvicorn mcp_audio_server.main:app --host 0.0.0.0 --port 8000
   

API Documentation

The full OpenAPI specification is available at /docs when the server is running:

http://localhost:8000/docs

Endpoints

Endpoint	Description
POST /analyze_chords	Analyze audio for chords, key, and tempo
GET /health	Health check endpoint
GET /ready	Readiness check for load balancers
GET /metrics	Prometheus metrics (redirects to metrics server)

Request Format

{
  "audio_data": "base64-encoded-audio-data",
  "format": "wav",  // "wav", "mp3", "ogg", "m4a", "flac"
  "options": {
    "model": "basic"  // "basic" or "advanced"
  }
}

Response Format

{
  "schema_version": "1.0.0",
  "key": "C",
  "tempo": 120.0,
  "chords": [
    {
      "time": 0.0,
      "label": "C",
      "confidence": 0.9
    }
  ],
  "duration": 4.0,
  "processing_info": {
    "sample_rate": 44100,
    "channels": 1,
    "processing_time": 0.245,
    "model_used": "basic"
  },
  "correlation_id": "550e8400-e29b-41d4-a716-446655440000"
}

Error Responses

{
  "error_code": "DECODING_ERROR",
  "message": "Failed to decode audio file: Invalid file format",
  "details": {
    "format": "wav",
    "error": "Invalid WAV header"
  },
  "correlation_id": "550e8400-e29b-41d4-a716-446655440000"
}

Development

See CONTRIBUTING.md for development guidelines.

Testing

# Run all tests with coverage reporting
pytest

# Run specific tests
pytest tests/analysis/test_tempo.py

Documentation

API documentation is generated from the FastAPI app and available at /docs.

To rebuild the documentation site:

# Install documentation dependencies
poetry install --with docs

# Build documentation
poetry run mkdocs build

# Serve documentation locally
poetry run mkdocs serve

License and Compliance

This project is licensed under the MIT License - see the LICENSE file for details.

Dependencies

All dependencies are documented in DEPENDENCIES.md with their respective licenses.

Changelog

See CHANGELOG.md for the project's version history and changes.