MoodBeats Emotion Based Music Recommendation

Music has a significant impact on human emotions and psychological well-being. However, selecting music that matches a person’s emotional state can be difficult, especially when mood fluctuates throughout the day. This project presents an Emotion-Based Music Recommendation System that automatically detects a user’s emotion and suggests music accordingly. The system captures facial expressions using a camera or accepts manual mood input. Using DeepFace, along with deep learning frameworks such as TensorFlow and Keras, the model identifies emotions like happiness, sadness, anger, fear, surprise, and excitement with high accuracy. The detected emotion is matched with pre-tagged music datasets, and suitable songs are recommended through Spotify or YouTube APIs. The user interface is developed using Streamlit, providing real-time interaction and playlist generation. Experimental results show an emotion detection accuracy of 85–92% and playlist relevance of 78–84%. The system enhances user experience by offering personalized playlists and can be applied in areas like mental wellness, fitness, and smart vehicles.