Skip to content Home Contact Mobile MyRMIT Library A-Z
RMIT UniversityResearch Repository
 

Emotion recognition in natural speech using empirical mode decomposition and Renyi entropy

He, L, Lech, M, Maddage, N and Allen, N 2009, 'Emotion recognition in natural speech using empirical mode decomposition and Renyi entropy', in 2009 International Symposium on Bioelectronics and Bioinformatics (ISBB2009), Netherlands, 9-11 Dec 2009, pp. 108-111.

Document type: Conference Paper
Collection: Conference Papers

Title Emotion recognition in natural speech using empirical mode decomposition and Renyi entropy
Author(s) He, L
Lech, M
Maddage, N
Allen, N
Year 2009
Conference name 2009 International Symposium on Bioelectronics and Bioinformatics (ISBB2009)
Conference location Netherlands
Conference dates 9-11 Dec 2009
Proceedings title 2009 International Symposium on Bioelectronics and Bioinformatics (ISBB2009)
Publisher Elsevier BV
Place of publication Netherlands
Start page 108
End page 111
Total pages 4
Abstract A new approach to the feature extraction process for automatic emotion classification in speech is presented and tested. The proposed feature extraction is based on the empirical mode decomposition (EMD) combined with the calculation of Renyi entropy. The proposed method was tested on natural speech data subjectively annotated with five different emotions: angry, anxious, dysphoric, happy and neutral. The data represented 44 male and 27 female speakers. Each emotion was represented by 200 utterances of an average duration 1.5 s. The modeling and classification was based on the Gaussian mixture model (GMM). The classification results for the Renyi entropy of order 2 produced an average correct classification rate of 48% varying only slightly across different emotions (std=7.5).
Subjects Signal Processing
ISSN 1746-8094
 
Versions
Version Filter Type
Access Statistics: 32 Abstract Views  -  Detailed Statistics
Created: Thu, 02 Jun 2011, 15:22:00 EST by Catalyst Administrator