FACEBEAT: An Emotion Based Music System Survey
M S Sahana, K V Sudheesh, B Gokul, K Likith, Prerana Thejraj, B S Raghuram · 2025
Through emotion recognition technology artificial intelligence and affective computing help build better music recommendation systems. These systems identify emotions in real-time by reading facial expressions and movement as well as checking physical and vocal signs. They then suggest music that fits the current mood or changes it to match desired emotions. The system detects emotions better to customize music in ways that users feel and respond to. Emotion-based music players work with machine learning to decode emotional data so users can easily change their music choices. This work analyzes multiple emotion reading methods such as facial expression scanning, text sentiment detection from speech and physical signals while examining their potential to aid music selection. The discussion includes emotions detection precision problems alongside privacy concerns and how users can shape their music collections. Technological advances enable better personal music experiences and support clinical treatment as well as entertainment applications. This research proposal is an emotion-driven music player that can recommend songs that match the user’s mood—happy, sad, neutral, or angry. The application will use the input from either heart rate data obtained from a smart band or facial images captured by a mobile camera. Two methods of classification are used: heart rate-based and facial image-based analysis. Once the emotion is detected, the application provides suggestions for songs that match the detected mood. Experimental results demonstrate the approach’s effectiveness, especially in accurately identifying the happy emotion due to its broader heart rate range.