• Version
• Download 9
• File Size 515.44 KB
• File Count 1
• Create Date 11 June 2025
• Last Updated 11 June 2025

A MUSIC RECOMMENDATION SYSTEM INTEGRATING FACIAL EXPRESSION RECOGNITION

Authors:

1. Latha¹, Dodla Aksthitha², Chedimela Sharon³, Danvath Abhijeet⁴,

Yaramadha Rishi Kumar Reddy⁵

^1-5 Department of CSE & TKR College of Engineering & Technology

^2-5B.Tech Students

ABSTRACT:

This project presents a deep learning-based music recommendation system that personalizes song suggestions based on the user's emotional state. Emotions such as happiness, sadness, anger, neutral, and surprise are detected using facial and hand landmarks extracted via MediaPipe's Holistic model. A TensorFlow/Keras classifier predicts emotions from these features, and the system recommends mood-matching songs from a curated dataset. Built with Flask, the application includes secure login, role-based access, and a user-friendly interface. Future work includes integration with real-time APIs like Spotify or YouTube Music for dynamic playlist generation and live streaming.

Keywords — Facial Landmarks, Mediapipe, TensorFlow, Keras, Computer Vision, Real-time Emotion Detection, Webcam-based Emotion Analysis.