Our faces contain a huge amount of information and it plays an important role in finding out of an individual‘s behavior. Facial expression analysis is the most effective way to detect human emotion. In this paper, we propose an approach for playing music and change the wallpaper for personal computer based on recognizing emotions through facial expressions displayed in live video streams. Facial expression recognition refers to the classification of facial features such as happiness, sadness, fear, disgust, surprise, and anger. To extract this type of feature we detect faces in front of a webcam we build own dataset according to facial expression and corresponding Haar classifiers are used. Then we make a playlist based on emotion and user interest and also build a model of the changing wallpaper in such a way that it will play songs fitting to our mood and changing our desktop wallpaper to happy images if we are sad. If we are angry it may change it to calming images.