EMOTION AWARE MUSIC RECOMMENDATION SYSTEM USING REAL TIME FACIAL EXPRESSION ANALYSIS

Abstract

Music plays an important role in influencing human emotions and mental well-being. Traditional music recommendation systems rely mainly on user preferences, listening history, or manual selection, which may not always reflect the user’s current emotional state. This project presents an Emotion Aware Music Recommendation System using Real Time Facial Expression Analysis that automatically detects the user’s mood and recommends suitable music accordingly. The proposed system utilizes a web-based application built with Python Flask as the backend and a browser-based interface for real-time user interaction. Facial images captured through the user’s webcam are processed and analyzed using the DeepFace emotion recognition framework, which identifies emotions such as happiness, sadness, anger, surprise, fear, neutral, and disgust from facial expressions. The detected emotion is then mapped to an appropriate set of songs stored in the system or retrieved from online platforms such as Spotify or YouTube, enabling personalized music recommendations. The system processes captured images by decoding base64 image data, performing facial emotion analysis, and returning recommended songs based on the dominant emotion detected. . Experimental evaluation demonstrates that the proposed system provides quick and context-aware music recommendations, improving user experience by aligning music suggestions with the user's real-time emotional state. This approach enhances traditional recommendation methods by incorporating affective computing and real-time emotion detection for intelligent and adaptive music selection.