Recognition of Emotions in German Speech Using Gaussian Mixture Models

0356050 - ÚFE 2011 RIV DE eng C - Conference Paper (international conference)
Vondra, Martin - Vích, Robert
Recognition of Emotions in German Speech Using Gaussian Mixture Models.
MULTIMODAL SIGNAL: COGNITIVE AND ALGORITHMIC ISSUES. Vol. 5398. Berlin: SPRINGER-VERLAG, 2009 - (Esposito, A.; Hussain, A.; Marinaro, M.; Martone, R.), s. 256-263. Lecture Notes in Artificial Intelligence, 5398. ISBN 978-3-642-00524-4. ISSN 0302-9743.
[euCognition International Training School on Multimodal Signals - Cognitive and Algorithmic Issues (European COST A2102). Vietri sul Mare (IT), 21.04.2008-26.04.2008]
R&D Projects: GA MŠMT OC08010
Institutional research plan: CEZ:AV0Z20670512
Keywords : emotion recognition * speech emotions
Subject RIV: JA - Electronics ; Optoelectronics, Electrical Engineering

The contribution describes experiments with recognition of emotions in German speech signal based oil the same principle as recognition of speakers. The most robust algorithm for speaker recognition is based On Gaussian Mixture Models (GMM). We examine three parameter Sets: the first contains suprasegmental features, in the second are segmental features and the last is a combination of the two previous parameter sets. Further we want to explore the dependency of the classification accuracy Oil the number of GMM model components. The aim of this contribution is a recommendation the number of GMM components and the optimal selection of speech parameters for emotion recognition in German speech.
Permanent Link: http://hdl.handle.net/11104/0194672