Two-stage Classification of Emotional Speech

Author

Xiao, Zhongzhe ; Dellandrea, Emmanuel ; Dou, Weibei ; Chen, Liming

Author_Institution

Dept. of MI, Ecole Centrale de Lyon, Ecully

fYear

2006

fDate

29-31 Aug. 2006

Firstpage

32

Lastpage

32

Abstract

The purpose of this paper is to make an automatic classification of speech into seven emotional classes as anger, boredom, disgust, fear, gladness, neutral and sadness. A two-stage classification composed of several sub-classifiers is proposed. A feature set with 68 features has been computed over 286 speech samples from the Berlin database. The sequential forward selection method (SFS) has been used for each classifiers of the two stages in order to decide the feature subsets in each step. The result for the first stage as three-state classification is 87%, and the global result of the seven emotional classes is 78%, where the correct recognition rate of random classification by chance is about 15%

Keywords

emotion recognition; feature extraction; speech recognition; automatic classification; emotional classes; emotional speech; feature set; random classification; sequential forward selection method; two-stage classification; Acoustic measurements; Electronic mail; Emotion recognition; Mel frequency cepstral coefficient; Motion pictures; Pediatrics; Spatial databases; Speech analysis; TV; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Digital Telecommunications, , 2006. ICDT '06. International Conference on

Conference_Location

Cote d´Azur

Print_ISBN

0-7695-2650-0

Type

conf

DOI

10.1109/ICDT.2006.76

Filename

1698479