EMOSIC: Emotion-Based Music Player Using Haar Cascade and Deep CNN

Abstract

This paper is aimed at enhancing the user experience of listening to music by incorporating emotion detection through facial recognition technology. Facial expressions are used to detect the user’s emotional state and the system recommends a playlist matching their mood. After each song, the emotion detection process is iterated to make sure that the next song played serves the user’s current emotional state. This is made possible by coining emotion detection algorithms and machine learning techniques. The user’s facial expressions are captured using a camera which is connected to the system, and the image is processed to identify the emotion using deep learning models. The system then matches the detected emotion with a pre-defined playlist that matches the user’s mood and plays a random song from that playlist. This creates a personalized music experience for the user. The proposed paper aims to provide users a unique and interactive music-listening experience. The system can accurately determine the user’s emotional state and provide them with music that matches their mood. The repetition of the emotion detection process after each song ensures that the playlist remains relevant to the user’s emotional state, creating a more personalized music experience.