Geodesic
A Novel Hierarchical Speech Emotion Recognition Method Based on Improved DDAGSVM
Computer Science and Information Systems, Tome 7 (2010) no. 1.

Voir la notice de l'article provenant de la source Computer Science and Information Systems website

Since error classification cumulation in hierarchical method impacts the classification ability of it, in this paper, a novel hierarchical method based on improved Decision Directed Acyclic Graph SVM (improved DDAGSVM) is proposed for speech emotion recognition. The improved DDAGSVM is constructed according to the confusion degrees of emotion pairs. In addition, a geodesic distance-based testing algorithm is proposed for the improved DDAGSVM to give the test samples differently distinguished many decision chances. Informative features and SVM optimized parameters used in each node of the improved DDAGSVM are gotten by Genetic Algorithm (GA) synchronously. On the Chinese Speech Emotion Database (CSED) and the Berlin Emotional Speech Database (BESD), the recognition experiment results reveal that, compared with multi-SVM, binary decision tree and traditional DDAGSVM, the improved DDAGSVM has the higher recognition accuracy with few selected informative features and moderate time for 7 emotions.
Keywords: Speech Emotion Recognition, Improved DDAGSVM, Hierarchical Recognition Method, Confusion Degree, Geodesic Distance 
@article{CSIS_2010_7_1_a18,
     author = {Qi-rong Mao and Yong-zhao Zhan},
     title = {A {Novel} {Hierarchical} {Speech} {Emotion} {Recognition} {Method} {Based} on {Improved} {DDAGSVM}},
     journal = {Computer Science and Information Systems},
     publisher = {mathdoc},
     volume = {7},
     number = {1},
     year = {2010},
     url = {http://geodesic.mathdoc.fr/item/CSIS_2010_7_1_a18/}
}
TY  - JOUR
AU  - Qi-rong Mao
AU  - Yong-zhao Zhan
TI  - A Novel Hierarchical Speech Emotion Recognition Method Based on Improved DDAGSVM
JO  - Computer Science and Information Systems
PY  - 2010
VL  - 7
IS  - 1
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/item/CSIS_2010_7_1_a18/
ID  - CSIS_2010_7_1_a18
ER  - 
%0 Journal Article
%A Qi-rong Mao
%A Yong-zhao Zhan
%T A Novel Hierarchical Speech Emotion Recognition Method Based on Improved DDAGSVM
%J Computer Science and Information Systems
%D 2010
%V 7
%N 1
%I mathdoc
%U http://geodesic.mathdoc.fr/item/CSIS_2010_7_1_a18/
%F CSIS_2010_7_1_a18
Qi-rong Mao; Yong-zhao Zhan. A Novel Hierarchical Speech Emotion Recognition Method Based on Improved DDAGSVM. Computer Science and Information Systems, Tome 7 (2010) no. 1. http://geodesic.mathdoc.fr/item/CSIS_2010_7_1_a18/