Determining efficiency of speech feature groups in emotion detection [Ses özni·teli·k gruplarinin duygu tespi·ti·nde etki·nli·kleri·ni·n beli·rlenmesi·]

Abstract

Features, extract from speech parameter are frequently used in emotion detection problem. Prosodic, MFCC, LPC and band energy feature groups are commonly used in literature to detect emotion in speech. The aim of the study is to examine the efficiency of these features groups in emotion detection problem using a SVM classifier.