Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS) dataset was used for implementation purposes of their model to segregate emotions. The proposed approach yields 81.17% accuracy.