The proposed system is a cutting-edge emotion-driven music recommendation system that uses machine learning and facial emotion recognition to dynamically choose and play music according to the user’s emotional state. The system intelligently creates customized playlists by utilizing Convolutional Neural Networks (CNNs) for real-time emotion analysis, which increases user satisfaction and engagement. It ensures a more engaging and emotionally impactful listening experience by responding to real-time emotional cues, in contrast to conventional recommendation models that depend on past preferences. Personalized music systems have advanced significantly with the combination of deep learning, computer vision, and dynamic playlist creation. This method makes music consumption more intuitive and therapeutic, and it may have uses beyond entertainment in mental health and emotional well-being. The study’s high accuracy (96.2%) in classifying emotions validates its efficacy and opens the door for further developments, such as wearable device integration, multimodal emotion recognition, and AI-driven personalization.