Detecting real life anger

Author

Burkhardt, Felix ; Polzehl, Tim ; Stegmann, Joachim ; Metze, Florian ; Huber, Richard

Author_Institution

Deutsche Telekom Labs., Berlin

fYear

2009

fDate

19-24 April 2009

Firstpage

4761

Lastpage

4764

Abstract

Acoustic anger detection in voice portals can help to enhance human computer interaction. A comprehensive voice portal data collection has been carried out and gives new insight on the nature of real life data. Manual labeling revealed a high percentage of non-classifiable data. Experiments with a statistical classifier indicate that, in contrast to pitch and energy related features, duration measures do not play an important role for this data while cepstral information does. Also in a direct comparison between Gaussian Mixture Models and Support Vector Machines the latter gave better results.

Keywords

Gaussian processes; acoustic signal processing; cepstral analysis; emotion recognition; human computer interaction; speech recognition; statistical analysis; support vector machines; Gaussian mixture models; acoustic anger detection; cepstral information; human computer interaction; real life anger; statistical classifier; support vector machines; voice portal data collection; voice portals; Acoustic signal detection; Automatic speech recognition; Automation; Classification algorithms; Clustering algorithms; Laboratories; Linear discriminant analysis; Portals; Support vector machine classification; Support vector machines; classification; detection; emotion; speech;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech and Signal Processing, 2009. ICASSP 2009. IEEE International Conference on

Conference_Location

Taipei

ISSN

1520-6149

Print_ISBN

978-1-4244-2353-8

Electronic_ISBN

1520-6149

Type

conf

DOI

10.1109/ICASSP.2009.4960695

Filename

4960695