🎹 Synthesizer Music Analysis Toolkit

A comprehensive, modular audio analysis toolkit designed specifically for composers and synthesizer music creators. This toolkit goes beyond academic metrics to provide intuitive, creative insights into your music using descriptors like "spacey," "organic," "crystalline," and "oozy."

✨ Features

🎵 Creative Mood Analysis

117 Creative Descriptors: Core (9), Extended (8), Advanced (100) - from spacey and organic to tempestuous, celestial, velvet, and microscopic
Character Classification: 59 tags across synthesis types, textures, and processing - from analog synths to crystalline textures to reverbed atmospheres
Phase-by-Phase Analysis: Detailed breakdown of each musical section with mood descriptors

🎯 Intelligent Song Sequencing

Automatic Playlist Generation: Creates optimal listening sequences based on musical flow principles
Key Compatibility: Uses circle of fifths relationships for smooth transitions
Energy Arc Management: Builds natural progression from atmospheric to climactic sections
Tempo Progression: Ensures gradual tempo changes for better flow

📊 Comprehensive Analysis

Musical Phase Detection: Automatically identifies intro, development, climax, and conclusion sections
Cluster Analysis: Groups similar tracks for themed playlists
Creative Insights: Understand your compositional patterns and sound palette
Multiple Export Formats: CSV, JSON, and Markdown reports optimized for further analysis

🏗️ Architecture

New Modular Design with Parallel Processing (v2.1)

The toolkit has been completely refactored into a professional Python package structure with comprehensive parallel processing capabilities:

audio_analysis/
├── __init__.py                 # Main package interface with parallel components
├── core/                       # Core analysis algorithms
│   ├── feature_extraction.py   # 80+ audio features with detailed comments
│   ├── feature_extraction_base.py  # 🆕 Shared feature extraction core
│   ├── parallel_feature_extraction.py  # 🆕 Parallel feature extraction
│   ├── phase_detection.py      # Musical structure detection
│   ├── clustering.py           # K-means clustering for track grouping
│   ├── parallel_clustering.py  # 🆕 Distributed clustering with tensor support
│   ├── tensor_operations.py    # 🆕 Hardware-agnostic tensor processing
│   └── sequencing.py           # Intelligent song ordering
├── analysis/                   # Creative interpretation modules
│   ├── mood_analyzer.py        # 117 creative mood descriptors
│   ├── character_analyzer.py   # 59 character tags (synthesis, texture, processing)
│   └── descriptors.py          # Comprehensive descriptor definitions and thresholds
├── utils/                      # Support utilities
│   ├── audio_io.py            # File loading and validation
│   ├── data_processing.py     # Data cleaning and standardization
│   ├── visualization.py       # Publication-quality plots
│   ├── type_conversion.py     # Centralized type conversion utilities
│   ├── validation.py          # Shared validation functions
│   └── statistics.py          # Statistical calculation utilities
├── exporters/                  # Output format handlers
│   ├── csv_exporter.py        # CSV export for spreadsheets
│   ├── json_exporter.py       # JSON for programmatic access
│   └── markdown_exporter.py   # Human-readable reports
├── api/                        # Main interfaces
│   ├── analyzer.py            # Primary AudioAnalyzer class
│   ├── parallel_analyzer.py   # 🆕 ParallelAudioAnalyzer for scalable processing
│   └── mcp_server.py          # FastMCP server integration
└── cli/                        # Command-line interface
    └── main.py                # Enhanced CLI with extensive options

Key Architectural Improvements (v2.1)

🚀 Parallel Processing:

6x+ Performance: Multi-core processing with automatic optimization
Tensor-Ready: Data structures optimized for hardware acceleration
Scalable: Configurable batch sizes and memory limits
Hardware-Agnostic: CPU, GPU, and future Tenstorrent processor support

🔧 Code Quality:

Single Source of Truth: Shared feature extraction core eliminates duplication
Consistent Results: Traditional and parallel processing produce identical outputs
Easy Extension: Add new moods and characters in one location
Better Testing: Comprehensive test coverage with shared utilities

⚡ Performance Benchmarks:

Feature Extraction: 6.7x speedup (8 cores)
Phase Detection: 5.6x speedup (8 cores)
Clustering: 5.0x speedup (8 cores)
Total Analysis: 6.1x speedup (8 cores)

🚀 Quick Start

Installation

# Clone or download the project
cd analyze_synths

# Create virtual environment
python -m venv .

# Activate virtual environment
source bin/activate  # On macOS/Linux
# or
.\Scripts\activate  # On Windows

# Install dependencies
pip install -r requirements.txt

Usage Options

🎵 Local Directory Analysis (Multiple Options)

Option 1: Wrapper Script (Easiest)

# Basic analysis - uses wrapper script
python analyze_library.py /path/to/your/music/directory

# With custom clustering
python analyze_library.py /path/to/music --clusters 3

# Export in specific format
python analyze_library.py /path/to/music --export-format markdown

# Get processing time estimate
python analyze_library.py /path/to/music --estimate

# Verbose output for debugging
python analyze_library.py /path/to/music --verbose

# Show supported formats and capabilities
python analyze_library.py --info

Option 2: Direct Module Access

# Same functionality, different invocation method
python -m audio_analysis.cli.main /path/to/your/music/directory

# All other options work the same way
python -m audio_analysis.cli.main /path/to/music --clusters 3
python -m audio_analysis.cli.main --info

Option 3: Parallel Processing Demo (NEW v2.1)

# Run parallel processing demonstration
python parallel_demo.py /path/to/music --workers 8 --batch-size 16

# Enable tensor optimizations for hardware acceleration
python parallel_demo.py /path/to/music --enable-tensor --device cpu

# Run specific demo modes
python parallel_demo.py /path/to/music --demo extraction    # Feature extraction only
python parallel_demo.py /path/to/music --demo clustering    # Clustering only
python parallel_demo.py /path/to/music --demo complete      # Full analysis

🌐 MCP Server Mode (Multiple Options)

Option 1: Wrapper Script (Easiest)

# Start MCP server - uses wrapper script
python mcp_server.py

# For use with FastMCP directly
fastmcp run mcp_server.py

Option 2: Via analyze_library.py wrapper

# Start MCP server with default settings
python analyze_library.py --mode mcp

# Start on custom host/port
python analyze_library.py --mode mcp --host 0.0.0.0 --port 8080

Option 3: Direct Module Access

# Direct module invocation
python -m audio_analysis.cli.main --mode mcp

# With custom settings
python -m audio_analysis.cli.main --mode mcp --host 0.0.0.0 --port 8080

🐍 Python API Usage

Standard Analysis:

from audio_analysis import AudioAnalyzer

# Initialize analyzer
analyzer = AudioAnalyzer('/path/to/audio/files')

# Run complete analysis
df = analyzer.analyze_directory()

# Perform clustering
cluster_labels, centers, features = analyzer.perform_clustering()

# Generate sequence recommendations  
sequence = analyzer.recommend_sequence()

# Export all results (default: all formats)
export_info = analyzer.export_comprehensive_analysis()

# Export specific formats
export_info = analyzer.export_comprehensive_analysis(export_format="markdown")
export_info = analyzer.export_comprehensive_analysis(export_format="json", base_name="my_analysis")

Parallel Processing (NEW v2.1):

from audio_analysis import ParallelAudioAnalyzer, ProcessingConfig

# Configure parallel processing
config = ProcessingConfig(
    max_workers=8,                    # Use 8 CPU cores
    batch_size=16,                    # Process 16 files per batch
    enable_tensor_optimization=True,  # Enable tensor operations
    memory_limit_mb=4096             # 4GB memory limit
)

# Initialize parallel analyzer
analyzer = ParallelAudioAnalyzer('/path/to/audio/files', config)

# Same interface as standard analyzer
df = analyzer.analyze_directory()
cluster_labels, centers, features = analyzer.perform_clustering()
sequence = analyzer.recommend_sequence()
export_info = analyzer.export_comprehensive_analysis(export_format="all")

# Get parallel processing statistics
stats = analyzer.get_processing_statistics()
print(f"Parallel speedup: {stats['parallel_processing_stats']['parallel_speedup']:.1f}x")
print(f"Throughput: {stats['parallel_processing_stats']['throughput']:.1f} files/second")

Hardware Acceleration (Future-Ready):

from audio_analysis import TensorFeatureExtractor

# CPU processing with tensor optimizations
extractor = TensorFeatureExtractor(device="cpu")
features = extractor.extract_features_from_paths(audio_file_paths)

# Tenstorrent processing (when available)
extractor = TensorFeatureExtractor(device="tenstorrent", device_id=0)
features = extractor.extract_features_from_paths(audio_file_paths)

Export Format Options (NEW v2.2):

from audio_analysis import AudioAnalyzer

analyzer = AudioAnalyzer('/path/to/audio/files')
analyzer.analyze_directory()

# Export all formats (default: CSV data + JSON + Markdown + visualizations)
analyzer.export_comprehensive_analysis()

# Export only specific formats for faster processing
analyzer.export_comprehensive_analysis(export_format="markdown")  # Human-readable report only
analyzer.export_comprehensive_analysis(export_format="json")     # Programmatic data only  
analyzer.export_comprehensive_analysis(export_format="csv")      # Spreadsheet data only

# Customize file naming
analyzer.export_comprehensive_analysis(
    export_format="json", 
    base_name="my_project"
)  # → Creates: my_project_data.json

# Full customization
analyzer.export_comprehensive_analysis(
    export_dir="/custom/path",
    export_format="markdown", 
    base_name="album_analysis",
    show_plots=True
)  # → Creates: album_analysis_comprehensive_report.md

📁 What You Get

Each analysis creates a timestamped directory with organized results:

📈 Data Files (CSV format)

audio_features.csv - Complete feature matrix (80+ features per track)
cluster_analysis.csv - Musical groupings and characteristics
phase_analysis.csv - Detailed phase breakdown with mood descriptors
sequence_recommendations.csv - Optimal track ordering with reasoning
summary_statistics.csv - High-level collection insights

🎨 Visualizations

phase_timeline.png - Visual timeline of musical phases for all tracks
cluster_analysis.png - Multi-panel cluster visualization with PCA
mood_distribution.png - Mood analysis across your collection
sequence_recommendations.png - Visual representation of optimal flow

📝 Reports

comprehensive_analysis_report.md - The main report with:
- Executive summary with key insights
- Recommended listening sequence with detailed reasoning
- Track-by-track mood and character analysis
- Phase-by-phase structural breakdown
- Cluster analysis for playlist creation
- Creative insights and compositional recommendations
analysis_data.json - Complete analysis data for programmatic access

🎼 Understanding Your Music

Mood Descriptors (117 Total)

Core Moods:

Spacey: Low energy, ethereal, expansive atmospheres
Organic: Natural textures, acoustic-like characteristics
Synthetic: Clean, precise, distinctly electronic
Oozy: Slow, flowing, liquid-like textures
Pensive: Contemplative, moderate energy, thoughtful
Tense: High energy, sharp, angular characteristics
Exuberant: Joyful, high energy, bright
Glitchy: Fragmented, stuttering, digital artifacts
Chaos: Extreme energy, unpredictable, intense

Extended Moods:

Ethereal: Delicate, floating, otherworldly
Atmospheric: Environmental, ambient, immersive
Crystalline: Clear, precise, bell-like
Warm: Comfortable, enveloping, intimate
Melodic: Tuneful, songlike, memorable
Driving: Forward-moving, rhythmic, propulsive
Percussive: Rhythmic, transient, beat-focused
Droning: Sustained, minimal, hypnotic

Character Tags (59 Total)

The expanded character system now includes three comprehensive categories:

Synthesis Types (25 total):

Analog Synth: Warm, vintage synthesizer characteristics
Digital Synth: Clean, precise digital synthesis
FM Synth: Complex harmonic interactions from frequency modulation
Granular Synth: Fragmented, particulate textures
Wavetable Synth: Morphing spectral evolution
Physical Modeling: Realistic acoustic behavior through physics simulation
Subtractive/Additive Synth: Filtered or harmonic-stacked synthesis
Modular Synth: Experimental, unpredictable characteristics
Pad/Lead/Bass/Arp Synth: Role-specific synthesizer voices
Piano/Organ/Choir/Brass/Woodwind: Acoustic instrument emulations
Drum Machine/Sampler: Rhythmic and sampling-based sources
Plus 10 additional specialized synthesis types

Texture Types (20 total):

Rich/Pure Texture: Complex layered vs. simple clean textures
Bright/Warm Harmonics: High vs. low frequency emphasis
Smooth/Rough Texture: Silky vs. gritty surface characteristics
Crystalline/Organic Texture: Glass-like precision vs. natural breathing
Mechanical/Liquid Texture: Industrial precision vs. flowing fluidity
Metallic/Wooden/Glassy: Material-inspired sonic characteristics
Fabric/Sandy/Rubbery: Tactile texture associations
Plus 8 additional texture descriptors

Processing Types (14 total):

Reverbed/Delayed: Spatial effects and echo patterns
Chorused/Flanged/Phased: Modulation-based effects
Distorted/Filtered: Harmonic saturation and frequency shaping
Compressed/Pitched: Dynamic control and frequency shifting
Ring Modulated: Metallic, inharmonic characteristics
Bit Crushed: Digital degradation and lo-fi character
Tape/Tube Saturated: Vintage warmth and harmonic distortion
Plus 5 additional processing characteristics

🎯 Creative Applications

For Composers

Study Your Patterns: Understand your compositional tendencies through data
Album Sequencing: Use AI-generated optimal listening sequences
Sound Palette Analysis: Identify your dominant musical characteristics
Structural Insights: Learn from phase analysis of your work
Mood Development: Track emotional progression in your compositions

For Producers

Playlist Creation: Use cluster analysis for themed collections
Remix Planning: Find compatible phases and sections across tracks
Energy Management: Plan DJ sets using detailed energy progression data
Mood Matching: Pair tracks with complementary emotional characteristics
Track Selection: Find similar tracks for consistent album flow

For Musicians

Practice Planning: Sequence practice sessions using energy flow principles
Performance Sets: Create dynamic live performance sequences
Collaboration: Share detailed analysis with band members and collaborators
Inspiration: Discover hidden patterns in your creative process
Learning: Understand how your music affects listeners emotionally

🛠 Technical Requirements

Python 3.8+ (Developed and tested on 3.13)
Audio Formats: WAV, AIFF, MP3 (WAV recommended for best quality)
Memory: 4GB+ RAM for large collections (100+ files)
Storage: Analysis exports typically 1-50MB per session
Dependencies: All handled by requirements.txt

📊 Sample Output

Command Line

$ python analyze_library.py /Users/composer/my_tracks

Initializing audio analyzer for: /Users/composer/my_tracks
Found 12 audio files. Processing...
Processing 1/12: ambient_dawn.wav
Processing 2/12: crystalline_patterns.wav
...

✓ CSV files exported to data/
✓ Plots saved to images/  
✓ Markdown report saved to comprehensive_analysis_report.md
✓ JSON results saved to analysis_data.json

============================================================
ANALYSIS COMPLETE
============================================================
Files processed: 12
Features extracted: 89
Phases detected: 47
Clusters created: 3
Export directory: /Users/composer/my_tracks/audio_analysis_20241213_143022

Sequence Recommendations

RECOMMENDED LISTENING SEQUENCE
==============================================================

1. ambient_dawn.wav
   atmospheric • analog_synth • 3:45 • 72 BPM • C
   Opening track - sets the mood with atmospheric atmosphere

2. crystalline_patterns.wav  
   crystalline • digital_synth • 4:20 • 85 BPM • G
   Early exploration - introduces digital_synth textures

3. organic_flow.wav
   organic • mellotron • 5:12 • 92 BPM • G  
   Core development - showcases organic at 92 BPM

🌐 MCP Server Tools

When running in MCP mode, six powerful tools are available for remote analysis:

🎭 `analyze_audio_mood`

Comprehensive mood and character analysis with confidence scores

Input: Audio files (base64 encoded)
Output: 117 mood descriptors, 59 character tags, musical metrics

🎵 `analyze_audio_phases`

Musical structure detection with mood analysis per section

Input: Audio files (base64 encoded)
Output: Detailed phase breakdown with timing and characteristics

🎼 `recommend_song_sequence`

AI-powered optimal listening sequence generation

Input: Multiple audio files (base64 encoded)
Output: Recommended order with detailed reasoning

🎯 `analyze_audio_clusters`

K-means clustering for musical similarity grouping

Input: Audio files and optional cluster count
Output: Cluster analysis with musical groupings and characteristics

🎨 `comprehensive_audio_analysis`

Complete analysis pipeline with all features

Input: Audio files and export format preference
Output: Full analysis with mood, phases, clustering, and sequencing

ℹ️ `get_supported_formats`

System capabilities and format information

Input: None
Output: Supported formats, descriptors, analysis capabilities

🔧 Advanced Features

🚀 Parallel Processing (NEW v2.1)

Multi-Core Processing: Automatic utilization of all available CPU cores
Configurable Batching: Process multiple files simultaneously with optimized memory usage
Performance Scaling: 6x+ speedup on multi-core systems
Hardware Acceleration: Tensor-optimized data structures for future acceleration
Memory Management: Intelligent memory usage with configurable limits

🏗️ Hardware Acceleration Ready

Tensor Operations: Data structures optimized for Tenstorrent processors
Device Abstraction: Hardware-agnostic interface supports CPU, GPU, and specialized processors
Batch Processing: Optimal utilization of parallel processing units
Memory Efficiency: Minimize data movement between processing units
Future-Proof: Easy integration with emerging hardware acceleration platforms

📊 Comprehensive Performance Monitoring

Processing Statistics: Detailed performance metrics and throughput analysis
Parallel Speedup: Real-time calculation of performance improvements
Memory Usage: Monitoring and optimization of memory consumption
Error Tracking: Comprehensive error handling and reporting
Benchmarking: Built-in performance benchmarking tools

🎯 Extensible Architecture

Single Source of Truth: Add new mood descriptors in one place
Consistent Analysis: Traditional and parallel processing produce identical results
Easy Extension: Simple framework for adding new creative descriptors
Modular Design: Clean separation of concerns for easy maintenance
Comprehensive Testing: Shared utilities ensure consistent behavior

Extensive Inline Documentation

Every analytical approach is thoroughly documented with:

Why this method: Explanation of creative relevance
How it works: Technical implementation details
Parameter choices: Justification for thresholds and settings
Musical context: Connection to composition and music theory

Robust Error Handling

File validation: Comprehensive format and content checking
Graceful degradation: Analysis continues even with problematic files
Memory management: Efficient processing of large collections
Progress reporting: Detailed feedback during long operations

Professional Export Options

CSV: Optimized for spreadsheet analysis and data science
JSON: Structured data for programmatic access and APIs
Markdown: Human-readable reports with musical insights
Visualizations: Publication-quality plots and charts

🔧 Troubleshooting

Common Issues

Out of Memory: Process files in smaller batches, increase system RAM
MP3 Issues: Install additional codecs, convert to WAV for best results
No Files Found: Verify audio files are in supported formats (WAV, AIFF, MP3)
Import Errors: Ensure all dependencies installed: pip install -r requirements.txt
MCP Server Issues: Install FastMCP: pip install fastmcp

Getting Help

# Show detailed help (any method works)
python analyze_library.py --help
python -m audio_analysis.cli.main --help

# Show format information  
python analyze_library.py --info
python -m audio_analysis.cli.main --info

# Estimate processing time
python analyze_library.py /path/to/files --estimate
python -m audio_analysis.cli.main /path/to/files --estimate

# Run with verbose output for debugging
python analyze_library.py /path/to/files --verbose
python -m audio_analysis.cli.main /path/to/files --verbose

# Parallel processing demo and help
python parallel_demo.py --help
python example_mood_extension.py  # Shows how to add new mood descriptors

Performance Optimization

# For large collections, use parallel processing
python parallel_demo.py /path/to/files --workers 16 --batch-size 32

# Enable tensor optimizations
python parallel_demo.py /path/to/files --enable-tensor --device cpu

# Monitor memory usage and adjust batch size
python parallel_demo.py /path/to/files --batch-size 8 --memory-limit 2048

📚 Next Steps

For New Users

Start with a small collection (5-10 tracks) to understand the output
Read the comprehensive report - Focus on the markdown file first
Try the recommended sequence - Play tracks in suggested order
Explore the visualizations - Understand your musical patterns
Experiment with clustering - Create themed playlists

For Advanced Users

Use the Python API - Integrate with your existing workflow
Customize analysis parameters - Adjust clustering and export options
Run MCP server - Enable AI assistant integration
Process large collections - Use batch processing techniques
Contribute improvements - The modular architecture welcomes enhancements

Integration Examples

With AI Assistants

# Example for Claude/ChatGPT integration
from audio_analysis import AudioAnalyzer

analyzer = AudioAnalyzer('/path/to/music')
results = analyzer.analyze_directory()

# Send results to AI for creative interpretation
mood_analysis = results[['filename', 'primary_mood', 'mood_descriptors']]
# "Please analyze these mood patterns and suggest creative directions..."

With Music Software

# Export data for DAW integration
analyzer.export_comprehensive_analysis(export_format='json')
# Import JSON into your DAW or music management software

Batch Processing

# Process multiple directories (using wrapper script)
for dir in /path/to/albums/*; do
    python analyze_library.py "$dir" --export-format csv
done

# Or using direct module access
for dir in /path/to/albums/*; do
    python -m audio_analysis.cli.main "$dir" --export-format csv
done

Transform your music analysis from academic metrics to creative insights. Perfect for composers who want to understand their art through an intuitive, musical lens. 🎶

New in v2.1: Comprehensive parallel processing capabilities with 6x+ performance improvements, hardware acceleration readiness for Tenstorrent processors, tensor-optimized data structures, and a refactored architecture that eliminates code duplication while maintaining full backward compatibility.

Previous v2.0: Complete modular refactor with extensive inline documentation, enhanced CLI, robust error handling, professional-grade architecture, and convenient wrapper scripts for easy access.

🌐 Hugging Face Deployment (NEW v2.2)

The toolkit now includes a complete Hugging Face Spaces deployment for web-based audio analysis, making it accessible to users worldwide without any installation required.

🚀 Web Interface Features

Easy Access: Upload audio files directly through your web browser for instant analysis

Multiple Analysis Types: Comprehensive analysis, mood-only, or phase detection only
Export Options: Download results in Markdown, JSON, or CSV formats
Public Demo: Share and demonstrate your audio analysis capabilities
No Installation: Works immediately in any web browser

📁 Deployment Structure

gradio/
├── app.py                 # Complete Gradio web interface  
├── requirements.txt       # HF-specific dependencies
├── README.md             # Model card with proper YAML frontmatter
└── CLAUDE.md             # Deployment-specific guidance

🎯 Deployment Options

Option 1: Hugging Face Spaces (Recommended)

Create new Space at hf.co/new-space
Choose "Gradio" SDK
Upload contents of gradio/ directory
Automatic deployment at https://huggingface.co/spaces/yourusername/spacename

Option 2: Hugging Face Model Repository

Upload complete Python package as model repository
Users install via pip install git+https://huggingface.co/username/repo.git
Include comprehensive documentation and examples

Option 3: PyPI + HF Community

Package for PyPI distribution (pip install audio-analysis-toolkit)
List as community resource on Hugging Face Hub

🎵 Web Analysis Capabilities

Supported Audio: WAV (recommended), AIFF, MP3 up to 100MB, 1s-30min duration Analysis Results: Same 17 mood descriptors, 9 character tags, and phase detection as desktop version Export Formats: Human-readable reports, structured JSON data, and spreadsheet-ready CSV files Real-time Processing: Immediate results with downloadable complete analysis

💡 Benefits for Users

Accessibility: No Python knowledge or installation required
Educational: Interactive exploration of audio analysis concepts
Professional: API-quality outputs for integration workflows
Gateway: Introduction to full desktop toolkit capabilities
Shareable: Public demos for collaboration and teaching

The web interface maintains full compatibility with all desktop analysis features while providing an intuitive, browser-based experience that makes advanced audio analysis accessible to a broader audience.

Name		Name	Last commit message	Last commit date
Latest commit History 17 Commits
audio_analysis		audio_analysis
gradio		gradio
reports		reports
.gitignore		.gitignore
CLAUDE.md		CLAUDE.md
LICENSE		LICENSE
PARALLEL_PROCESSING.md		PARALLEL_PROCESSING.md
README.md		README.md
README_ENVIRONMENTS.md		README_ENVIRONMENTS.md
README_PERFORMANCE_COMPARISON.md		README_PERFORMANCE_COMPARISON.md
TENSTORRENT_INTEGRATION_MEMO.md		TENSTORRENT_INTEGRATION_MEMO.md
VIRTUAL_ENVIRONMENT_FIX_SUMMARY.md		VIRTUAL_ENVIRONMENT_FIX_SUMMARY.md
activate_standard.sh		activate_standard.sh
activate_tenstorrent.sh		activate_tenstorrent.sh
analyze_library.py		analyze_library.py
example_mood_extension.py		example_mood_extension.py
fix_virtualenv.sh		fix_virtualenv.sh
installed_packages.txt		installed_packages.txt
mcp_client_example.py		mcp_client_example.py
mcp_server.py		mcp_server.py
optimizations.txt		optimizations.txt
parallel_demo.py		parallel_demo.py
performance_comparison.py		performance_comparison.py
pyvenv.cfg		pyvenv.cfg
requirements.txt		requirements.txt
test_performance_demo.py		test_performance_demo.py
test_tenstorrent_integration.py		test_tenstorrent_integration.py

Folders and files

Latest commit

History

Repository files navigation

🎹 Synthesizer Music Analysis Toolkit

✨ Features

🎵 Creative Mood Analysis

🎯 Intelligent Song Sequencing

📊 Comprehensive Analysis

🏗️ Architecture

New Modular Design with Parallel Processing (v2.1)

Key Architectural Improvements (v2.1)

🚀 Quick Start

Installation

Usage Options

🎵 Local Directory Analysis (Multiple Options)

🌐 MCP Server Mode (Multiple Options)

🐍 Python API Usage

📁 What You Get

📈 Data Files (CSV format)

🎨 Visualizations

📝 Reports

🎼 Understanding Your Music

Mood Descriptors (117 Total)

Character Tags (59 Total)

🎯 Creative Applications

For Composers

For Producers

For Musicians

🛠 Technical Requirements

📊 Sample Output

Command Line

Sequence Recommendations

🌐 MCP Server Tools

🎭 analyze_audio_mood

🎵 analyze_audio_phases

🎼 recommend_song_sequence

🎯 analyze_audio_clusters

🎨 comprehensive_audio_analysis

ℹ️ get_supported_formats

🔧 Advanced Features

🚀 Parallel Processing (NEW v2.1)

🏗️ Hardware Acceleration Ready

📊 Comprehensive Performance Monitoring

🎯 Extensible Architecture

Extensive Inline Documentation

Robust Error Handling

Professional Export Options

🔧 Troubleshooting

Common Issues

Getting Help

Performance Optimization

📚 Next Steps

For New Users

For Advanced Users

Integration Examples

With AI Assistants

With Music Software

Batch Processing

🌐 Hugging Face Deployment (NEW v2.2)

🚀 Web Interface Features

📁 Deployment Structure

🎯 Deployment Options

🎵 Web Analysis Capabilities

💡 Benefits for Users

About

Resources

License

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

🎭 `analyze_audio_mood`

🎵 `analyze_audio_phases`

🎼 `recommend_song_sequence`

🎯 `analyze_audio_clusters`

🎨 `comprehensive_audio_analysis`

ℹ️ `get_supported_formats`

Packages