HomeBrowseUpload
← Back to registry
// Skill profile

Chord Analyzer Skill

name: chord-analyzer

by ctwww · published 2026-03-22

开发工具数据处理
Total installs
0
Stars
★ 0
Last updated
2026-03
// Install command
$ claw add gh:ctwww/ctwww-chord-analyzer
View on GitHub
// Full documentation

---

name: chord-analyzer

description: "Analyze music audio files to extract chord progressions, key signature, tempo, and song structure. Use when user wants to identify chords, analyze a song's harmony, or extract musical information from audio files (mp3, wav, m4a, etc.)."

homepage: https://github.com/librosa/librosa

metadata: { "openclaw": { "emoji": "🎸", "requires": { "bins": ["python3"], "pip": ["librosa", "numpy"] } } }

---

# Chord Analyzer Skill

Analyze music audio files to extract chord progressions, key signature, tempo, and song structure.

When to Use

✅ **USE this skill when:**

  • User wants to analyze a song's chords and harmony
  • "What are the chords in this song?"
  • "Analyze this audio file"
  • "Extract the chord progression"
  • "What key is this song in?"
  • User provides an audio file path and asks for musical analysis
  • When NOT to Use

    ❌ **DON'T use this skill when:**

  • Only wants general music info (lyrics, artist) → use web search
  • Wants to generate music → use music generation skills
  • Needs professional-grade transcription → recommend specialized software (Chordify, Hookpad)
  • Requires detailed instrument separation → use dedicated source separation tools
  • Supported Formats

  • **Audio**: mp3, wav, m4a, flac, ogg
  • **Duration**: Works best for songs under 5 minutes
  • Installation

    First time use requires installing dependencies:

    pip3 install librosa numpy scipy scikit-learn soundfile

    Usage

    Basic Analysis

    # Analyze an audio file
    python3 chord_analyzer.py
    
    # Edit the script to change the audio path
    # Default: /Users/chentiewen/Music/网易云音乐/example.mp3

    Script Integration

    Copy the `chord_analyzer.py` script to your workspace and modify the `audio_path` variable:

    audio_path = "/path/to/your/song.mp3"
    result = analyze_audio(audio_path)

    Output

    The analyzer provides:

    1. **Key Signature**: Detected musical key (e.g., C, F#m, G)

    2. **Tempo**: Speed in BPM with rhythm classification

    3. **Chord Progression**: Complete chord sequence with timestamps

    4. **Chord Statistics**: Most frequently used chords

    5. **Song Structure**: Intro/Verse/Outro segmentation (basic)

    Sample Output

    调性: F#m
    速度: 123.0 BPM
    节奏: 快板 (Allegro)
    
    和弦走向:
    F#mdim → A → D → Bm → E → A → D → Bm → E ...
    
    主要和弦:
      A: 15次 (20.3%)
      E: 14次 (18.9%)
      D: 12次 (16.2%)

    How It Works

    1. **Load Audio**: Uses `librosa.load()` to read audio at 22.05kHz

    2. **Extract Chroma**: Computes chroma features (pitch class profiles) using STFT

    3. **Detect Key**: Analyzes chroma energy across all 12 keys (major + minor)

    4. **Track Tempo**: Uses `librosa.beat.beat_track()` for tempo detection

    5. **Analyze Chords**: Samples chroma at measure boundaries and matches against chord templates

    6. **Merge & Simplify**: Combines consecutive identical chords

    Limitations

  • **Accuracy**: Chord detection is approximated; not professional-grade
  • **Complexity**: Struggles with heavily layered or distorted music
  • **Structure**: Simple segmentation (not verse/chorus detection)
  • **Melody**: Does not extract melodic lines or instrument parts
  • **Chord Extensions**: Detects basic triads (major, minor, diminished), not 7th/9th chords
  • For Complete Transcription

    For professional music transcription, recommend:

  • **Chordify**: https://chordify.net (online chord detection)
  • **Hookpad**: https://www.hooktheory.com/hookpad (theory + chords)
  • **MuseScore**: https://musescore.org (manual transcription)
  • **Capo**: https://capoapp.com (slow down + chord detection)
  • Notes

  • Analysis takes ~10-30 seconds depending on song length
  • Best results with clear, non-distorted audio
  • Works best for pop/rock/folk styles with clear harmony
  • Not suitable for atonal, experimental, or heavily percussive music
  • // Comments
    Sign in with GitHub to leave a comment.
    // Related skills

    More tools from the same signal band