Music Mood Matcher

An intelligent music recommendation system that analyzes your mood through natural language and suggests personalized playlists using advanced AI and machine learning techniques.

Comment

AI Music Mood Matcher - Interactive Demo

An intelligent music recommendation system that analyzes your mood through natural language and suggests personalized playlists using advanced AI and machine learning techniques.

Demo Features

Intelligent Mood Detection

  • Natural Language Processing: Describe your mood in plain English
  • Multiple AI Methods:
    • OpenAI GPT integration for advanced understanding
    • NLTK VADER sentiment analysis
    • Custom keyword-based detection
  • Real-time Analysis: Instant mood classification with confidence scores

Smart Music Recommendations

  • AI-Enhanced Song Analysis: ChatGPT analyzes individual songs for mood compatibility
  • Intelligent Song Selection: AI evaluates track characteristics beyond simple audio features
  • Dual Recommendation Modes:
    • Standard algorithmic matching
    • Advanced AI-powered curation (with OpenAI API key)
  • Personalized Explanations: Get detailed AI reasoning for each song recommendation
  • Song Comparison: AI ranks multiple songs for specific moods
  • Interactive Feedback: Rate recommendations to improve future suggestions

Comprehensive Analytics

  • Mood Distribution: Visualize your emotional patterns over time
  • Music Feature Analysis: Understand the audio characteristics of your preferences
  • Recommendation Insights: See why certain songs were suggested
  • Interactive Charts: Explore your data with dynamic visualizations

Quick Start

1. Clone & Setup

git clone https://github.com/djleamen/music-mood-matcher-demo
cd music-mood-matcher-demo

2. Install Dependencies

pip install -r requirements.txt

3. Run the Demo

# Option 1: Direct launch
streamlit run demo.py

# Option 2: Using start script
chmod +x start_demo.sh
./start_demo.sh

4. Open in Browser

Navigate to http://localhost:8501 to access the interactive demo.

System Requirements

  • Python: 3.8 or higher
  • Memory: 2GB RAM minimum
  • Storage: 100MB for dependencies

How to Use the Demo

Home Page

  • Overview of the system capabilities
  • Quick stats about the demo music library
  • Navigation to different features

Mood Matcher

  1. Enter Your Mood: Type how you're feeling in natural language

    • Examples: "I need energizing music for my workout"
    • "I want something chill for studying"
    • "I'm feeling nostalgic and want emotional songs"

  2. AI Analysis: Watch as the system:

    • Analyzes your text for sentiment and keywords
    • Determines your primary mood with confidence scores
    • Maps your emotion to audio features

  3. Get Recommendations: Receive a personalized playlist with:

    • 20 carefully selected tracks
    • Detailed explanations for each recommendation
    • Audio feature visualizations

  4. Provide Feedback: Rate songs to improve future recommendations

Advanced Configuration

OpenAI Integration (Optional)

For enhanced mood detection using ChatGPT:

  1. Get an OpenAI API key from OpenAI Platform
  2. Copy .env.example to .env
  3. Add your API key: bash OPENAI_API_KEY=your_api_key_here

Benefits of OpenAI Integration:

  • More nuanced mood understanding
  • Better handling of complex emotional descriptions
  • Contextual reasoning for recommendations
  • Improved accuracy for ambiguous inputs

Technical Architecture

Project Structure

musicanalyzer-demo/
├── demo.py                 # Main Streamlit application
├── core/                   # Core AI modules
│   ├── __init__.py
│   ├── mood_analyzer.py    # NLP and mood detection
│   ├── music_analyzer.py   # Audio feature analysis
│   └── recommendation.py   # Recommendation engine
├── requirements.txt        # Python dependencies
├── start_demo.sh          # Launch script
├── .env.example           # Environment template
└── README.md

AI Components

Mood Analyzer

  • NLTK VADER: Sentiment polarity analysis
  • TextBlob: Additional sentiment processing
  • OpenAI GPT: Advanced natural language understanding

Enhanced AI Recommendation Engine

  • ChatGPT Song Analysis: Individual song evaluation for mood compatibility
  • Intelligent Song Selection: AI considers track name, artist, and audio features
  • Contextual Understanding: Incorporates user's specific mood description
  • Comparative Ranking: AI compares multiple songs and ranks them for specific moods
  • Detailed Explanations: Provides reasoning for each song recommendation
  • Dual-Mode Operation: Seamlessly switches between AI and algorithmic approaches

AI Features Available with OpenAI API Key

  1. Enhanced Mood Detection: More accurate understanding of complex emotions
  2. Smart Song Curation: AI analyzes each song individually for optimal mood matching
  3. Personalized Explanations: Get detailed reasoning for why specific songs were chosen
  4. Song Comparison Tool: AI ranks multiple songs for any mood category
  5. Contextual Recommendations: AI considers your specific mood description when selecting music

Music Analyzer

  • Audio Features: Analyzes danceability, energy, valence, tempo, etc.
  • Clustering: Groups similar tracks using K-means
  • Feature Scaling: Normalizes audio characteristics
  • Correlation Analysis: Finds patterns in music preferences

Recommendation Engine

  • Cosine Similarity: Matches tracks to mood profiles
  • Weighted Scoring: Balances multiple recommendation factors
  • Diversity Injection: Ensures playlist variety
  • Feedback Learning: Adapts to user preferences over time

Machine Learning Pipeline

  1. Text Processing: Clean and tokenize mood descriptions
  2. Feature Extraction: Convert text to numerical representations
  3. Mood Classification: Predict emotional state from input
  4. Audio Mapping: Match moods to musical characteristics
  5. Recommendation Scoring: Rank tracks by relevance
  6. Diversity Optimization: Select varied but relevant tracks

Demo Scenarios

Try these example inputs to explore different features:

🏃‍♂️ Workout Energy

"I need high-energy music for an intense workout session"

Expected Output: High-energy electronic, rock, and hip-hop tracks

📚 Study Focus

"I want calm, instrumental music for deep concentration"

Expected Output: Ambient, classical, and lo-fi tracks

💔 Emotional Processing

"I'm going through a breakup and need songs that understand my pain"

Expected Output: Emotional ballads and introspective tracks

🎉 Party Vibes

"I'm hosting a party and need upbeat danceable music"

Expected Output: Pop, dance, and party classics

🌙 Late Night Chill

"It's 2 AM and I want something dreamy and atmospheric"

Expected Output: Chillout, downtempo, and atmospheric tracks

Future Enhancements

Potential Features

  • Real Spotify Integration: Connect to your actual music library
  • Voice Input: Speak your mood instead of typing
  • Multiple Languages: Support for non-English mood descriptions
  • Social Features: Share playlists and mood insights
  • Smart Scheduling: Automatic mood detection based on time/calendar

Technical Improvements

  • Deep Learning: Neural networks for better mood understanding
  • Real-time Audio: Analyze currently playing music
  • Collaborative Filtering: Learn from community preferences
  • Cross-platform: Desktop and mobile applications

Learning Resources

AI/ML Concepts Used

  • Natural Language Processing: Text analysis and sentiment detection
  • Machine Learning: Classification and recommendation algorithms
  • Data Science: Feature engineering and statistical analysis
  • Web Development: Interactive application design

Technologies Explored

  • Python: Core programming language
  • Streamlit: Web application framework
  • Scikit-learn: Machine learning library
  • NLTK: Natural language processing
  • Plotly: Interactive data visualization
  • OpenAI: Advanced AI integration

Acknowledgments

  • OpenAI: GPT integration for advanced NLP
  • Spotify: Audio feature definitions and inspiration
  • Streamlit: Excellent web framework for ML demos
  • NLTK: Robust natural language processing tools
  • Scikit-learn: Comprehensive machine learning library

🎵 Ready to explore your musical emotions? Launch the demo and discover how AI can understand your mood through music!

streamlit run demo.py

Built With

Share this project:

Updates