Speech emotion recognition using spiking neural networks

Abstract

Human social communication depends largely on exchanges of non-verbal signals, including non-lexical expression of emotions in speech. In this work, we propose a biologically plausible methodology for the problem of emotion recognition, based on the extraction of vowel information from an input speech signal and on the classification of extracted information by a spiking neural network. Initially, a speech signal is segmented into vowel parts which are represented with a set of salient features, related to the Mel-frequency cesptrum. Different emotion classes are then recognized by a spiking neural network and classified into five different emotion classes. © Springer-Verlag Berlin Heidelberg 2006.