Please use this identifier to cite or link to this item: https://doi.org/10.1016/j.csl.2013.06.002
Title: Speaker state classification based on fusion of asymmetric simple partial least squares (SIMPLS) and support vector machines
Authors: Huang, D.-Y.
Zhang, Z.
Ge, S.S. 
Keywords: Asymmetric SIMPLS
Fusion
Partial least squares
Sleepiness detection
Speaker state recognition
Speech emotion recognition
Support Vector Machine
Issue Date: Mar-2014
Citation: Huang, D.-Y., Zhang, Z., Ge, S.S. (2014-03). Speaker state classification based on fusion of asymmetric simple partial least squares (SIMPLS) and support vector machines. Computer Speech and Language 28 (2) : 392-414. ScholarBank@NUS Repository. https://doi.org/10.1016/j.csl.2013.06.002
Abstract: This paper presents our studies of the effects of acoustic features, speaker normalization methods, and statistical modeling techniques on speaker state classification. We focus on the investigation of the effect of simple partial least squares (SIMPLS) in unbalanced binary classification. Beyond dimension reduction and low computational complexity, SIMPLS classifier (SIMPLSC) shows, especially, higher prediction accuracy to the class with the smaller data number. Therefore, an asymmetric SIMPLS classifier (ASIMPLSC) is proposed to enhance the performance of SIMPLSC to the class with the larger data number. Furthermore, we combine multiple system outputs (ASIMPLS classifier and Support Vector Machines) by score-level fusion to exploit the complementary information in diverse systems. The proposed speaker state classification system is evaluated with several experiments on unbalanced data sets. Within the Interspeech 2011 Speaker State Challenge, we could achieve the best results for the 2-class task of the Sleepiness Sub-Challenge with an unweighted average recall of 71.7%. Further experimental results on the SEMAINE data sets show that the ASIMPLSC achieves an absolute improvement of 6.1%, 6.1%, 24.5%, and 1.3% on the weighted average recall value, over the AVEC 2011 baseline system on the emotional speech binary classification tasks of four dimensions, namely, activation, expectation, power, and valence, respectively. © 2013 Elsevier Inc. All rights reserved.
Source Title: Computer Speech and Language
URI: http://scholarbank.nus.edu.sg/handle/10635/83045
ISSN: 08852308
DOI: 10.1016/j.csl.2013.06.002
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.

SCOPUSTM   
Citations

15
checked on Jun 14, 2018

WEB OF SCIENCETM
Citations

12
checked on May 1, 2018

Page view(s)

54
checked on Jun 8, 2018

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.