Emotion Sound Classification with Support Vector Machine Algorithm

Abstract

Speech one of the biometric characteristic owned by human being, as well as fingerprint, DNA, retina of the eyes and so not the two human beings who have the same voice. Human emotion is a matter that can only be predicted through the face of a person, or from the change of facial expression but it turns out human emotions can also be detected through the spoken voice. Someone emotion are happy, angry, neutral, sad, and surprise can be detected through speech signal. The development of voice recognition system is still running at this moment. So that ini this research, the analysis of someone emotion through speech signal. Some related research about the sound aims to have process of identity recognition gender recognition, Emotion recognition based on conversation. In this research the writer does research on the emotional classification of speech two classes started from happy, angry, neutral, sad and surprise while the used algorithm in this research is SVM (Support Vector Machine) with alghoritmMFCC (Mel-frequency cepstral coefficient)for extraction where it contains filter process that adapted to human’s listening. The result of the implementation process of both algorithms gives accuracy level ashappy=68.54%, angry=75.24%, neutral=78.50%, sad=74.22% and surprise=68.23%.