In this present competitive world people are so busy and engaged in many works and in that process they are undergoing through various emotions like Sad ,Happy, Anger etc. In order to cope up with the varying emotions the best activity that people choose is listening to music. Being able to suggest a song or music playlist enhances the listening experience because music elicits strong emotions and feelings in people. There exist some music recommendation systems in order to recommend the music to the users we use techniques such as Collaborative filtering and models such as Context-Based Information Retrieval system. These early models have some limitations such as dynamic evolvement, playlist generation and they are not personalized systems. In this Paper it is proposed to develop a prototype to endorse dynamic music system based on human emotions. The main objectives for this paper is to propose a model for Mood derivation of the user using Haar Cascade Algorithm and endorsing the music to the user based on the derived mood using facial landmark extraction.

Face Detection, Open CV, Dynamic Music, Human Emotion, Landmark Extraction