Human emotion assumes a fundamental part lately. Feeling depends on human sentiments which can be both communicated or not. Feelings communicate the human's singular way of behaving which can be in various structures. Extraction of the emotion states people individual condition of conduct. The goal of this task is to separate element from human face and identify feeling and to play music as indicated by the emotion identified. In this task, fostering a model to suggest dynamic music proposal framework in view of human feelings is our principal perspective. From the genuine face the emotion is recognized and when the mind-set is gotten from input picture, particular melodies for the state of mind would be played to hold the clients. In this methodology, the application gets associated with human sentiments accordingly giving an individualized touch to the clients. Our projected framework focuses on distinguishing the human face using Haar Cascade classifier. A few stages like face location utilizing webcam, facial feature extraction, look identification, with that interfacing with a music player will be followed. For exploratory outcomes, Haar cascade calculation and Computer vision along with CNN (Convolution Neural Network) for feeling discovery and music suggestion.