🎧 Moodify: An Emotion-Based Music Recommender

Moodify is a real-time emotion-based music recommendation system that uses facial expression analysis to recommend songs that match the user's current emotional state. By leveraging a Convolutional Neural Network (CNN) trained on the FER2013 dataset and integrating with the Spotify API, Moodify personalizes music experiences dynamically as your emotions change.

---

🎯 Project Highlights

• 🎥 Real-time facial emotion detection using webcam feed
• 🧠 Deep Learning model (CNN) trained on the FER2013 dataset
• 🎵 Spotify API integration for emotion-aligned music recommendations
• 🌐 Flask web app with live video stream and dynamic song suggestions
• 📊 Detailed evaluation with accuracy, confusion matrix, ROC curve, and more

---

📸 Demo

Live detection and instant playlist recommendations.

---

🛠 Technologies Used

Technology	Purpose
Python	Core programming language
TensorFlow/Keras	Deep learning model for emotion classification
OpenCV	Real-time face detection and image preprocessing
Spotipy	Python client for Spotify Web API
Flask	Web application framework
Pandas, NumPy	Data processing and handling
Matplotlib	Data visualization and evaluation metrics

---

📂 Project Structure

Moodify/
├── app.py                   # Main Flask app
├── camera.py                # Handles video capture and emotion detection
├── train.py                 # Model training script
├── Spotipy.py               # Spotify playlist fetch script
├── emotion_detection_model.keras  # Trained model file
├── songs/                   # Emotion-wise CSV playlists
├── templates/               # HTML files (index.html)
├── static/                  # CSS and JS assets
├── haarcascade_frontalface_default.xml  # Face detection XML
├── utils.py                 # Supporting utility functions

---

📊 Dataset Used

• FER2013: Facial Expression Recognition 2013
  • Images: 35,887 grayscale images (48x48)
  • Classes: Angry, Disgust, Fear, Happy, Neutral, Sad, Surprise
  • Split: 28,709 for training, 7,178 for testing

---

🚀 Setup Instructions

1. Clone the Repository

git clone https://github.com/TheDevilPython/Moodify.git
cd Moodify

2. Create Virtual Environment

python -m venv moodify-env
source moodify-env/bin/activate  # or moodify-env\Scripts\activate on Windows

3. Install Dependencies

pip install -r requirements.txt

4. Set Environment Variables

Add your Spotify credentials:

set SPOTIPY_CLIENT_ID=your_client_id
set SPOTIPY_CLIENT_SECRET=your_client_secret

5. Run the App

python app.py

---

📈 Model Training (Optional)

To retrain the model:

python train.py

It includes:

• Data augmentation
• Accuracy/loss visualization
• Early stopping
• Evaluation metrics like confusion matrix, ROC curve

---

📑 Key Features

• Real-time facial expression capture
• CNN-based emotion classification
• Spotify API integration using Spotipy
• Dynamic web UI to show current mood and playlist
• Downloadable CSVs of songs for each mood
• Fully modular and extensible

---

📷 Screenshots

Live Emotion Detection	Song Recommendations

---

📌 Future Improvements

• Add support for multi-modal emotion detection (voice + text)
• Improve playlist personalization based on past emotion history
• Add speech-based interaction
• Deploy via Docker or on cloud (Heroku, AWS)

---

📄 License

This project is licensed under the MIT License. Feel free to use and modify.

---

👨‍💻 Author

Made with ❤️ by [Your Name]
GitHub: TheDevilPython