Speech Emotion Recognition System

Abstract

Speech Emotion Recognition (SER) is a research topic which has a wide range of applications. The features of speech such as, Mel Frequency cepstrum coefficients (MFCC ) are extracted which are uttered in the speech. To classify different emotional states such as boredom, happiness, sadness, neutral, anger, from various emotional sound tracks from a database containing emotional speech SVM is used as classifier. SVM is used for classification of emotions. The accuracy obtained in SVM is very much higher.
DOI: 10.17762/ijritcc2321-8169.150573