A Deep Learning-Based Framework for Mood-Oriented Music Recommendation Using Facial Expression Analysis

Computer Engineering

Music plays a significant role in influencing human emotions, behavior, and mental well-being. Research has shown that music can positively affect neurological functions and emotional states, making it an essential component of daily life. With the rapid growth of digital music platforms, users increasingly prefer personalized music recommendations that align with their current emotions and moods. This study presents an intelligent mood-based music recommendation system that utilizes computer vision and deep learning techniques to automatically suggest songs based on a user's facial expressions. The proposed framework captures facial images through an integrated camera and employs facial expression analysis to determine the user's emotional state. Face detection is performed using the Haar Cascade algorithm, while emotion classification is achieved through a Convolutional Neural Network (CNN) model. Based on the detected emotion, the system generates personalized music recommendations in real time, eliminating the need for manual song selection. The use of an inbuilt camera not only improves system efficiency but also reduces hardware costs compared to traditional sensor-based solutions. This paper discusses the design, implementation, and performance evaluation of the proposed framework. Experimental results demonstrate that CNN-based facial emotion recognition can effectively identify user moods and provide relevant music recommendations, thereby enhancing user engagement, satisfaction, and overall listening experience.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.