DEEP LEARNING DRIVEN EMOTION-AWARE MUSIC RECOMMENDATION FRAMEWORK
DEEP LEARNING DRIVEN EMOTION-AWARE MUSIC RECOMMENDATION FRAMEWORK
Authors:
Ganteda Manoj Kumar 1, M. Sathwika 2
1 Assistant Professor, 2 MCA Final Semester, Master of
Computer Applications, Sanketika Vidya Parishad Engineering College, Vishakhapatnam,
Andhra Pradesh, India
gantedamanojkumar@gmail.com,muilisathwika33@gmail.com
ABSTRACT
In the modern digital world, personalization has become a key factor in improving user experience, especially in entertainment applications like music streaming. Music is closely connected to human emotions and plays a major role in influencing mood, relaxation, and mental well-being. However, most traditional music recommendation systems rely only on user listening history, ratings, and preferences, which limits their ability to provide accurate recommendations based on the user’s current emotional state.To overcome this limitation, the Emotion-Based Music Recommendation System is proposed. This system uses facial emotion recognition to detect the user’s real-time mood and suggest songs accordingly. It integrates Artificial Intelligence (AI), Machine Learning (ML), and Computer Vision techniques to create an intelligent and user-centric application.The system captures facial expressions through a webcam and processes them using OpenCV. A Convolutional Neural Network (CNN) model built with TensorFlow/Keras is used to classify emotions such as happy, sad, angry, neutral, and surprised. Based on the detected emotion, the system filters a music dataset using pandas and displays suitable song recommendations through a Flask-based web interface.