Skip to main content
Glama
mcp-audio-analysis

mcp-audio-analysis

by hugohow

MCP Music Analysis

This repository contains a Model Context Provider (MCP) that uses MCP and librosa for audio analysis on audio in local, youtube link, or audio link.

Usage with Claude Desktop

Related MCP server: multimodal-reader-mcp

Installation

Installing via Smithery

To install Music Analysis for Claude Desktop automatically via Smithery:

npx -y @smithery/cli install @hugohow/mcp-music-analysis --client claude

Manual Installation

# Clone repository
git clone git@github.com:hugohow/mcp-music-analysis.git
cd mcp-music-analysis

# Create virtual environment and install
uv venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate
uv pip install -e .

Usage with Claude Desktop

Locate Configuration File

The configuration file location depends on your operating system:

  • macOS:

    ~/Library/Application\ Support/Claude/claude_desktop_config.json
  • Windows:

    %APPDATA%\Claude\claude_desktop_config.json
  • Linux:

    ~/.config/Claude/claude_desktop_config.json

Add the following to your claude_desktop_config.json:

{
  "mcpServers": {
    "music-analysis": {
      "command": "uvx",
      "args": ["-n", "mcp-music-analysis"]
    }
  }
}

Example Prompts

Here are some sample prompts you might use in a conversational or chat-based context once the server is running. The MCP will understand these requests and execute the relevant tools:

Can you analyze the beat of /Users/hugohow-choong/Desktop/sample-6s.mp3?
Could you give me the duration of https://download.samplelib.com/mp3/sample-15s.mp3 ?
Please compute the MFCC for this file: /path/to/another_audio.mp3
What are the spectral centroid values for /path/to/music.wav?
I'd like to know the onset times for https://www.youtube.com/watch?v=8HFiFd9vx1c

To-Do List

  • Add URL to audio file download

  • Add YouTube to audio file transformation

  • Experiment with multiple Python environments (testing)

  • Improve installation guide

  • Integrate Whisper for lyrics

  • Implement a Docker solution

Author

Hugo How-Choong

Available Tools

8 tools
beat_trackB
Computes the beat track of the given audio time series using librosa.
The beat track is a representation of the audio signal in terms of its
rhythmic content, which is useful for music analysis.
The beat track is computed using the following parameters:
- hop_length: The number of samples between frames.
- start_bpm: The initial estimate of the tempo (in BPM).
- tightness: The tightness of the beat tracking (default is 100).
- units: The units of the beat track (default is "frames"). It can be frames, samples, time.

Inputs: path_audio_time_series_y (required), hop_length, start_bpm, tightness, +1 more.

chroma_cqtA
Computes the chroma CQT of the given audio time series using librosa.
The chroma CQT is a representation of the audio signal in terms of its
chromatic content, which is useful for music analysis.
The chroma CQT is computed using the following parameters:
- path_audio_time_series_y: The path to the audio time series (CSV file).
    It's sometimes better to take harmonics only
- hop_length: The number of samples between frames.
- fmin: The minimum frequency of the chroma feature.
- n_chroma: The number of chroma bins (default is 12).
- n_octaves: The number of octaves to include in the chroma feature.
The chroma CQT is saved to a CSV file with the following columns:
- note: The note name (C, C#, D, etc.).
- time: The time position of the note in seconds.
- amplitude: The amplitude of the note at that time.
The path to the CSV file is returned.

Inputs: path_audio_time_series_y (required), hop_length, fmin, n_chroma, +1 more.

download_from_urlC

Downloads a file from a given URL and returns the path to the downloaded file. Be careful, you will never know the name of the song.

Inputs: url (required).

download_from_youtubeC

Downloads a file from a given youtube URL and returns the path to the downloaded file. Be careful, you will never know the name of the song.

Inputs: youtube_url (required).

get_durationC

Returns the total duration (in seconds) of the given audio time series.

Inputs: path_audio_time_series_y (required).

loadB

Loads an audio file and returns the path to the audio time series Offset and duration are optional, in seconds. Be careful, you will never know the name of the song.

Inputs: file_path (required), offset, duration.

mfccC
Computes the MFCC of the given audio time series using librosa.
The MFCC is a representation of the audio signal in terms of its
spectral content, which is useful for music analysis.
The MFCC is computed using the following parameters:
- path_audio_time_series_y: The path to the audio time series (CSV file).
    It's sometimes better to take harmonics only

Inputs: path_audio_time_series_y (required).

tempoC

Estimates the tempo (in BPM) of the given audio time series using librosa. Offset and duration are optional, in seconds.

Inputs: path_audio_time_series_y (required), hop_length, start_bpm, std_bpm, +2 more.

Tool Schema Changelog

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

Loading tool schema changelog…

    TDQS

    B3.1/5.0

    Scored across 8 tools

    Disambiguation4/5

    Most tools have distinct purposes in audio analysis (e.g., beat tracking vs. chroma vs. MFCC), but download_from_url and download_from_youtube overlap significantly in function—both download audio files from different sources with similar descriptions. The core analysis tools are well-differentiated, but the download tools could cause confusion.

    Naming Consistency3/5

    The naming is mixed: some tools use descriptive verb_noun patterns (beat_track, get_duration), while others are single verbs (load, tempo) or noun phrases (chroma_cqt, mfcc). There's no consistent convention across all tools, though most names are readable and hint at their function.

    Tool Count4/5

    With 8 tools, the count is reasonable for an audio analysis server, covering key operations like loading, downloading, and various analysis features. It's slightly lean but not incomplete, as it includes essential functions without being overwhelming.

    Completeness3/5

    The server covers basic audio analysis tasks (loading, downloading, duration, tempo, spectral features), but there are notable gaps: no tools for editing, filtering, or exporting audio, and no way to visualize results. The analysis tools are focused on computation but lack broader workflow support.

    Related MCP Connectors

    Related MCP Servers