Skip to content

Latest commit

 

History

History
20 lines (16 loc) · 1.56 KB

README.md

File metadata and controls

20 lines (16 loc) · 1.56 KB

Project Title: SoulTunes - Emotion-Based Music Recommendation System

Project Overview:

Developed a real-time emotion-based music recommendation system using CNNs and OpenCV for facial emotion detection, providing personalized music playlists.

Key Features:

  1. Integrated Haar Cascade and CNN for real-time facial emotion recognition, mapping emotions like happiness, sadness, and anger to relevant music.
  2. Built a web-based interface using HTML, CSS, and JavaScript, allowing users to receive music recommendations based on detected emotions.
  3. Developed the system using Python, TensorFlow, and Keras, training the model on the Kaggle dataset for accurate emotion analysis.
  4. Integrated the platform with Spotify, using the API to stream personalized music based on emotional states.

Key Challenges:

  1. Improved emotion detection accuracy, especially for complex emotions like anger and fear, using advanced CNN training.
  2. Maintained real-time performance for emotion detection and music recommendations using efficient OpenCV processing.
  3. Addressed privacy concerns related to facial recognition data by implementing privacy-compliant practices.

Learning Outcomes:

  1. Gained expertise in CNNs and OpenCV for emotion detection and model training.
  2. Enhanced skills in building full-stack applications with a focus on user experience and emotion-based personalization.
  3. Overcame technical challenges in linking real-time emotion detection to a dynamic music recommendation system.