Machine Learning Techniques based Audio Player to Soothe Human Emotions

People in the current world are suffering from lot of stress related diseases due to various reasons. High stress levels may lead to various health hazards like high blood pressure, heart attack etc. One of the stress relief activities is listening to music. If the music played does not suit the current emotion of the listener, it may aggravate stress of the user further. Emotion based music player is a music player based on machine learning techniques which suggests the songs of the playlist based on person's emotions. This paper proposes an emotion-based music player, which suggests songs based on user's various emotions namely happy, sad, angry and neutral. The application captures the user's photo through web camera and processes the facial image to identify user's emotion using machine learning techniques. Based on the emotion detected, it selects some song to play. The proposed application is more accurate in determining human emotion than existing techniques.