Facial Emotion Based Automatic Music Recommender System

Listening to music can change the mood of any individual based on type of song like classic, pop or hip-hop. Music can have very strong influence on an individual's emotions. These emotions can be recognized by extracting various facial image features like Eye or Lip movement. Using facial expressions, meaning and context of words can be clearly understood by Humans. With the quick improvement of versatile systems and advanced interactive media innovations, advanced music has gotten to be the standard customer substance looked for by numerous youthful people. People frequently utilize music for mood regulation, more specifically to change a bad mood, to get motivated, to focus on their work, to increase energy levels or to reduce tension. Moreover, tuning in to the correct kind of music at the proper time may progress mental wellbeing. Recommendation of different songs based on an individual's current facial expressions could be latest research problem within the field of image processing. Very few researchers have proposed the solution of this research problem through automatic facial recognition using convolution neural network (CNN) models. In this paper, the proposed system based on CNN model focused on detecting human emotions from face image to identify list of songs to be played through music player to swing user's mood. The objective of this study is to improve the accuracy and reduce time and cost of proposed system for music recommendation.