For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Pathological Voice Detection Using Efficient Combination of Heterogeneous Features
Ji-Yeoun LEE Sangbae JEONG Minsoo HAHN
IEICE TRANSACTIONS on Information and Systems
Publication Date: 2008/02/01
Online ISSN: 1745-1361
Print ISSN: 0916-8532
Type of Manuscript: LETTER
Category: Speech and Hearing
pathological voice detection, heterogeneous feature combination, mel-frequency filter bank energies, higher-order statistics, pattern classification algorithm,
Full Text: PDF>>
Combination of mutually complementary features is necessary to cope with various changes in pattern classification between normal and pathological voices. This paper proposes a method to improve pathological/normal voice classification performance by combining heterogeneous features. Different combinations of auditory-based and higher-order features are investigated. Their performances are measured by Gaussian mixture models (GMMs), linear discriminant analysis (LDA), and a classification and regression tree (CART) method. The proposed classification method by using the CART analysis is shown to be an effective method for pathological voice detection, with a 92.7% classification performance rate. This is a noticeable improvement of 54.32% compared to the MFCC-based GMM algorithm in terms of error reduction.