Speaker recognition in emotional environment

Abstract

This paper deals with development of speaker recognition system in emotional environments. In this paper, Mel-frequency cepstral coefficients (MFCC) have been used to represent the speaker specific information. A simulated speech corpus of Hindi language are used to check the performance of speaker recognition in emotional environment. The emotions included in this study are anger, neutral, sad, happy and surprise. Emotion recognition models are developed using Gaussian mixture models. Performance of speaker recognition is studied in emotional environment. The results show that emotions play vital role in speaker recognition. © 2012 Springer-Verlag.