Music Recommendation Using Facial Emotions

Music lets us feel all the possible emotions that we experience in our lives. Music can help raise someone's mood and also help people calm and relax themselves. The thought behind the topic was that the system utilizes the emotional expression of the user and recommends music that aligns with the emotions. This research paper explores the uses facial expression analysis and machine learning to provide personalized music recommendations. It uses emotional AI to understand and interpret users' facial expressions, allowing the system to gauge their mood and suggest songs that align with their feelings. The system uses OpenCV for facial recognition and TensorFlow for emotion detection. Once the user's emotional state is identified, it uses the Spotify API to curate a list of songs that match the detected emotions. This seamless integration ensures users receive timely and appropriate music recommendations. Initial tests have shown high accuracy in detecting emotions, with positive feedback on the relevance of the recommended songs. This innovative use of technology not only showcases the potential for personalized entertainment but also sets a new standard for user interaction in digital music platforms. As the technology continues to evolve, further improvements and refinements are expected to make personalized music recommendations even more accurate and enjoyable. The findings of this research aim to contribute to the develop features to assist music therapists in treating patients with mental stress, anxiety, acute depression, and trauma. Increased user engagement with the system, as personalized recommendations align more closely with individual tastes which will foster a sense of connection with the music content. The project will also result in a recommendation system that adopts to changes in user preferences over time. The system will help user in boosting their mood and have more optimistic state of mind.