🎵 Melodic Metadata Massacrer (MMM)

"In the symphony of digital rights, we are the conductors of chaos." 🎼⚡

MMM is a Python CLI tool that performs lossless removal of ALL watermarks and metadata from MP3 and WAV audio files, effectively thwarting systems that detect AI-generated music through embedded identifiers.

🎭 Features

Core Capabilities

• Complete Metadata Annihilation: Removes ID3, RIFF INFO, FLAC tags, and custom chunks
• AI Watermark Detection: Identifies spread spectrum, echo-based, and statistical watermarks
• Spectral Cleaning: Advanced frequency-domain watermark removal
• Fingerprint Elimination: Normalizes AI-generated statistical patterns
• Paranoid Mode: Maximum destruction with multiple cleaning passes
• Batch Processing: Parallel processing of entire directories
• Verification Engine: Before/after comparison with forensic reporting

Detection Methods

• Spread spectrum watermarks
• Echo-based signatures
• Statistical pattern analysis
• Phase modulation detection
• Amplitude modulation analysis
• Frequency domain anomalies

🚀 Quick Start

Installation

# Clone the repository
git clone https://github.com/research/mmm.git
cd mmm

# Create virtual environment with Python 3.10-3.13 (librosa doesn't support 3.14+ yet)
python3.10 -m venv mmm_env
source mmm_env/bin/activate  # On Windows: mmm_env\Scripts\activate

# Install the package and all dependencies
pip install -e .

Important: MMM requires Python 3.10-3.13 due to librosa dependencies. Python 3.14+ is not yet supported by numba/librosa.

GPU Acceleration (Optional - Recommended!)

For maximum performance (700x+ speedup), install GPU packages:

# Install GPU acceleration packages (NVIDIA GPU required)
pip install cupy-cuda12x torch torchaudio

# Verify GPU detection
python -c "import torch; print(f'CUDA: {torch.cuda.is_available()}')"

GPU Requirements:

• NVIDIA GPU with CUDA support (tested on RTX 3080 Ti)
• 4GB+ VRAM recommended
• CUDA 12.x compatible drivers

Performance Benchmarks

With GPU acceleration enabled (RTX 3080 Ti):

• 790x real-time processing speed
• 47,409 audio-minutes per minute throughput
• 0.27 seconds to analyze 214.6-second audio file
• Parallel processing across all CPU cores + GPU

Without GPU acceleration (CPU-only):

• Standard processing speed (slower for large files)
• Still utilizes multi-core CPU processing

Basic Usage

# Obliterate metadata from a single file
mmm obliterate dystopian_symphony.mp3

# Paranoid mode with verification
mmm obliterate suspicious_music.wav --paranoid --verify -o clean_output.wav

# Batch process directory
mmm massacre /path/to/music --paranoid --workers 8

# Analyze file without modifying
mmm analyze questionable_track.mp3

🔧 Configuration

MMM uses YAML configuration files for customization:

# ~/.config/mmm/config.yaml
paranoia_level: medium
preserve_quality: high
watermark_detection:
  - spread_spectrum
  - echo_based
  - statistical
output_format: preserve_original
backup_originals: true

Presets

• stealth: Maximum paranoia, quality preservation
• fast: Quick processing, basic cleaning
• quality: Preserve maximum audio quality
• research: Deep analysis, detailed logging

# Use preset
mmm config preset stealth

# Create custom preset
mmm config create my_preset --paranoid maximum --quality high

🎯 Commands

obliterate

Complete sanitization of individual files

mmm obliterate INPUT_FILE [OPTIONS]

Options:
  -o, --output PATH     Output file path
  --paranoid           Maximum destruction mode
  --verify             Verify watermark removal
  --backup             Create backup of original
  --format FORMAT      Output format (preserve/mp3/wav)
  --turbo              Enable GPU acceleration (700x+ faster)

massacre

Batch processing of directories

mmm massacre DIRECTORY [OPTIONS]

Options:
  -d, --output-dir PATH  Output directory
  -e, --extension TEXT   File extensions (multiple)
  -w, --workers INT      Parallel workers
  --paranoid            Paranoid mode
  --backup              Create backups

analyze

Deep forensic analysis without modification

mmm analyze INPUT_FILE                # Regular CPU mode
mmm analyze INPUT_FILE --turbo        # GPU accelerated mode (700x+ faster)

config

Configuration management

mmm config              Show current config
mmm config preset NAME  Apply preset

🎛️ Advanced Stealth Flags

These are opt-in, fine-grained toggles for research tuning. Defaults keep audio quality high; enable selectively:

• --gated-resample-nudge/--no-gated-resample-nudge (default off): ultra-tiny resample up/down applied only on higher-energy segments (minimal audibility, good stealth).
• --phase-noise/--no-phase-noise (default on): tiny FFT phase noise.
• --phase-swirl/--no-phase-swirl (default on): light all-pass swirl.
• --phase-dither/--no-phase-dither (default on), --comb-mask/--no-comb-mask, --transient-shift/--no-transient-shift: earlier experimental steps (may affect audio; leave off unless testing).
• --masked-hf-phase/--no-masked-hf-phase (default off): HF-only masked phase noise.
• --micro-eq-flutter/--no-micro-eq-flutter (default off): RMS-gated, <0.013 dB band flutter.
• --hf-decorrelate/--no-hf-decorrelate (default off): decorrelate only 12–16 kHz band.
• --refined-transient/--no-refined-transient (default off): ultra-small, onset-gated shifts.
• --adaptive-transient/--no-adaptive-transient (default off): onset-strength adaptive micro-shifts (~0.03–0.08 ms) with light blending.

Recommended “stealth” starting point (quality-preserving):

python -m mmm.cli obliterate input.mp3 -o output.mp3 --turbo --paranoid \
  --gated-resample-nudge --phase-noise \
  --no-phase-dither --no-comb-mask --no-transient-shift \
  --no-phase-swirl --no-masked-hf-phase --no-resample-nudge \
  --no-hf-decorrelate --no-micro-eq-flutter --no-refined-transient

Preset shortcut (applies the above flags): stealth-plus

# ~/.config/mmm/config.yaml (example)
preset: stealth-plus

You can also load it via mmm config preset stealth-plus (creates/uses preset file under your config dir).

Preset includes advanced flags:

• phase_dither=False, comb_mask=False, transient_shift=False
• phase_swirl=False, masked_hf_phase=False, resample_nudge=False
• gated_resample_nudge=True, phase_noise=True
• micro_eq_flutter=False, hf_decorrelate=False
• refined_transient=False, adaptive_transient=False

Notes on pattern suppression counts:

• “Patterns Found/Suppressed” in sanitization results come from the spectral cleaner’s suppression actions (e.g., attenuating suspicious bands/patterns) and do not imply detector-verified watermarks unless the detector reports them.
• Verification threat counts include metadata/container anomalies and detector findings; if the detector reports zero watermarks, remaining threats are likely metadata/binary anomalies rather than confirmed watermarks.

🧪 Detector Notes (Research)

We test against third-party detectors to understand robustness (not to guarantee evasion). Recent results:

• SubmitHub / SHLabs: “Inconclusive / mixed characteristics” with clean audio using --gated-resample-nudge --phase-noise and other advanced flags off. Temporal scores improved significantly; spectral scores became “could be AI / human unlikely”.
• Aggressive stacks (phase dither / comb mask / transient shift) degraded audio; not recommended.

Always audition audio locally before running external checks.

🛡️ Legal & Ethical Notice

⚠️ IMPORTANT: This tool is designed exclusively for authorized security research and educational purposes.

• Use only on files you own or have explicit permission to modify
• You are responsible for compliance with applicable laws and terms of service
• The developers do not condone or support copyright infringement
• This tool demonstrates vulnerabilities in watermarking systems for research purposes

📊 Technical Details

Architecture

┌─────────────────────────────────────────┐
│                CLI Layer                │
│  Click-based interface with personality  │
└─────────────────┬───────────────────────┘
                  │
┌─────────────────▼───────────────────────┐
│            Core Processing              │
│  • AudioSanitizer main engine          │
│  • FileProcessor for batch operations   │
└─────────────────┬───────────────────────┘
                  │
┌─────────────────▼───────────────────────┐
│          Detection Modules              │
│  • WatermarkDetector                   │
│  • MetadataScanner                     │
│  • StatisticalAnalyzer                  │
└─────────────────┬───────────────────────┘
                  │
┌─────────────────▼───────────────────────┐
│         Sanitization Modules            │
│  • MetadataCleaner                     │
│  • SpectralCleaner                     │
│  • FingerprintRemover                  │
└─────────────────────────────────────────┘

Dependencies

• Core: Click, NumPy, SciPy
• Audio: Librosa, PyDub, SoundFile
• Metadata: Mutagen
• UI: Rich, Colorama

Quality Preservation

MMM prioritizes audio quality while removing watermarks:

• Signal-to-Noise Ratio: >40dB target
• Quality Loss: <5% maximum
• Spectral Integrity: >90% preservation
• Dynamic Range: Preserved when possible

🧪 Development

Running Tests

# Install test dependencies
pip install pytest pytest-cov

# Run tests
pytest tests/ -v --cov=mmm

# Run specific test
pytest tests/test_audio_sanitizer.py -v

Contributing

1. Fork the repository
2. Create feature branch (git checkout -b feature/amazing-feature)
3. Commit changes (git commit -m 'Add amazing feature')
4. Push to branch (git push origin feature/amazing-feature)
5. Open Pull Request

Code Style

# Install development dependencies
pip install black flake8 mypy

# Format code
black mmm/

# Lint code
flake8 mmm/

# Type checking
mypy mmm/

📈 Performance Metrics

Operation	File Size	Processing Time	Quality Loss
MP3 (5 min)	5MB	~15s	<2%
WAV (5 min)	50MB	~30s	<1%
Paranoid Mode	Variable	2-3x slower	Same

Results on Intel i7-9700K, 16GB RAM

🔬 Research Applications

MMM is designed for academic and security research:

• Watermark Vulnerability Assessment: Test resilience of audio watermarking systems
• Privacy Research: Study audio fingerprinting and tracking
• Educational: Demonstrate audio steganography techniques
• Security Auditing: Verify effectiveness of watermark removal

🎨 CLI Experience

MMM features a unique hacker-aesthetic interface:

┌─────────────────────────────────────────────┐
│  ♪♫ MELODIC METADATA MASSACRER v2.0 ♫♪    │
│     "Making AI detectors cry since 2025"    │
└─────────────────────────────────────────────┘

🔍 Scanning: dystopian_symphony.mp3
😈 Found 47 metadata tags... time to DELETE THEM ALL!
🌊 Spectral watermark detected at 19.2kHz... NEUTRALIZING...
⚡ Statistical fingerprint obliterated!
✨ File sanitized! Your AI overlords will never know... 🤫

📝 License

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

🤝 Acknowledgments

• Open-source audio processing community
• Security researchers in digital watermarking
• Python audio processing ecosystem

⚡ Disclaimer

"The best place to hide a dead body is page two of the search results. But your metadata won't even make it there."

---

Remember: With great audio comes great responsibility. Use wisely. 🎼💀