Study on Optimization of Pathological Voice Features

DOI

10.2991/ameii-15.2015.123How to use a DOI?

Keywords

classification; optimal selection; pathological voice; recognition

Abstract

This paper proposes a method to optimize Pathological Voice Features based on classification and experimental design. Firstly, all voice parameters of features are classified, then pick the right amount of each class of representative and typical parameters based on the principle of mathematical statistics. Finally, recognition experiments are performed with this combination. C4.5 decision tree algorithm is used to recognize vocal cord or non-vocal cord damage voice, results show that recognition rate of optimized features is 5% higher than non-optimized, which is up to 89%.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Xiaojun Zhang
AU  - Yating Shao
AU  - Baoyin Sun
AU  - Zhi Tao
PY  - 2015/04
DA  - 2015/04
TI  - Study on Optimization of Pathological Voice Features
BT  - Proceedings of the International Conference on Advances in Mechanical Engineering and Industrial Informatics
PB  - Atlantis Press
SP  - 665
EP  - 669
SN  - 2352-5401
UR  - https://doi.org/10.2991/ameii-15.2015.123
DO  - 10.2991/ameii-15.2015.123
ID  - Zhang2015/04
ER  -