Vinícius J. D. Vieira, Rafael R. Pertum, Renato Candido. Sobre a influência de distorções acústicas na classificação de patologias laríngeas. Simpósio Brasileiro de Telecomunicações – SBrT’2024, 2024, Belém, Anais do SBrT 2024, 2024, pp. 1–5. (in portuguese)

Abstract

In this work, an analysis is carried out on the influence of acoustic variations in the classification of laryngeal pathologies. In this context, it is observed the behavior of the following characteristics extracted from voice signals: MFCC (Mel-Frequency Cepstral Coefficients), LPC (Linear Predictive Coefficients), F0 (fundamental frequency), Jitter and Shimmer. The acoustic variations considered are: different sampling rates, different noise levels, and different reverberation levels. The pathologies considered are: Reinke’s edema, carcinoma, leukoplakia, polyps and vocal fold paralysis. The classification stage is performed with quadratic discriminant analysis. The results indicate that the measurements present different accuracy behaviors according to the acoustic distortion. The MFCC measure was the one that presented the greatest robustness in all variation scenarios, considering a performance above 70\% in terms of accuracy, sensitivity and specificity.

Keywords

Speech signal processing, laryngeal pathologies classification, acoustic distortions, acoustic characteristics.

Downloads

• PDF