Music Recommender Based on the Facial Emotion of the User Identified Using YOLOV8

This research seeks to leverage the profound influence of music on an individual's emotions to enhance the quality and overall experience of listening to music. By uniting facial emotion recognition with music recommendation, this research not only advances the field of human-computer interaction but also holds promise for applications in entertainment, mental health, and beyond. The core innovation within our approach lies in the utilization of the cutting-edge YOLOv8 algorithm, which is the latest release in 2023 and comes equipped with advanced capabilities for tasks such as detection, segmentation, and classification. The proposed model detects the facial emotion of an individual and suggests music based on the detected emotion. The YOLOv8 model's performance on the FER-2013 and FER+ datasets, achieved an accuracy of 60.142% and 77.898%, respectively. Our study acknowledges dataset imbalances and proposes strategies for improving recognition accuracy for specific emotions. This paper aims to bridge the gap between facial emotion recognition and music recommendation, providing users with personalized music experiences tailored to their emotional states, unlike traditional preference-based music recommendations.