Aspect-Based Music Recommendation System Using CNN

Artificial intelligence has come a long way in the detection of emotions and expressions, especially through the analysis of users' facial expressions. The system needs to obtain live feeds from the device camera or external cameras in order to accomplish this. It must then analyze many aspects of human faces, including color, posture, expression, orientation, and illumination. Recent studies have looked into how Computer Vision (CV) and Machine Learning (ML) can be used to create systems that can identify different moods or emotions in people. Machine learning (ML) provides a number of methods for detecting emotions, allowing machines to learn how to recognize and comprehend various emotional states. People from different markets, ages, backgrounds, linguistic groups, musical tastes, political philosophies, and socioeconomic levels can come together via music. Because they can be used everywhere and integrated into daily activities like travel, sports, and more, music players and streaming apps are very popular. Whether it's to boost en- ergy levels, ease tension, or lift a negative mood, many individuals utilize music as a tool to regulate their mood. Studies have also indicated that mental wellbeing may be improved by listening to the right music at the right time. Emotions and music are therefore intimately intertwined. Our suggested method attempts to provide a mood-based music player that, in addition to detecting the user's mood in real-time and making appropriate song recommendations accessible on the web. Users receive personalized recommendations from e-commerce giants like Amazon and eBay based on their preferences and past purchases, while businesses like Spotify and Pandora utilize machine literacy. Deep literacy techniques for providing useful advice.