Nonlinear Dynamic Analysis of Pathological Voices
Research on the human health evaluation through sound analysis is now attracting more and more researchers in the world. Acoustic analysis could be a useful tool to diagnose the disease. Therefore, pathological voices can be used to evaluate the health status as a complementary technique, such as bronchitis. In this article, we proposed a nonlinear dynamic method to analysis pathological voices. Firstly, pathological voices were preprocessed and numerous features were extracted. Secondly, a binary coded chromosome genetic algorithm (GA) was applied as feature selection method to optimize feature descriptor set. The experimental results show that GA, PCA along with support vector machine (SVM) has the best performance in the pathology voices diagnosis.
KeywordsGA SVM PCA Pathology voice
Unable to display preview. Download preview PDF.
- 1.Fang, C.Y., Li, H.F., Ma, L., Hong, W.X.: Status and Development of Human Health Evaluation Based on Sound Analysis. In: First International Conference on Cellar Molecular Biology Biophysics and Bioengineering, p.66 (2010)Google Scholar
- 2.Fang, C.Y., Li, H.F.: Sound Analysis for Diagnosis of Children Health Based on MFCCE and GMM. In: International Review on Computers and Software, pp. 1153–1156 (2011)Google Scholar
- 7.Arjmandi, M.K.: Identifies Voice Disorders Using Long-time Features And Support Vector Machine With Different Feature Reduction Methods. Journal of Voice (2011)Google Scholar
- 8.Nayak, J., Bhat, P.S.: Identification of voice disorders using speech samples. IEEE Trans. 37, 951–953 (2003)Google Scholar
- 10.Fu, W.J., Yang, X.H., Wang, Y.T.: Heart Sound Diagnosis Based on DTW and MFCC. In: 2010 3rd International Congress on Image and Signal Processing, p. 2920 (2010)Google Scholar
- 13.Gavriely, N., Airflow, D.W.: Effects on Amplitude And Spectral Content of Normal Breath Sounds. Journal of Applied Physiology 80(1), 5–13 (1996)Google Scholar
- 15.Yang, J., Ames, I.A., Honavar, V.: Feature Subset Selection Using A Genetic Algorithm. Intelligent Systems and Their Applications, 44–49 (1998)Google Scholar
- 17.Bu, H.L., Zheng, S.Z., Xia, J.: Genetic Algorithm Based Semi-feature Selection Method. In: 2009 International Joint Conference on Bioinformatics Systems Biology and Intelligent Computing, pp. 521–524 (2009)Google Scholar
- 19.Eyben, F., Wóllmer, M., Schuller, B.: OpenSMILE - The Munich Versatile and Fast Open-Source Audio Feature Extractor. In: Proc. ACM Multimedia, pp. 1459–1462. ACM, Florence (2010)Google Scholar
- 20.The INTERSPEECH 2012 Speaker Trait Challenge, http://emotion-research.net/sigs/speech-sig/is12-speaker-trait-challenge