Vocal fold pathology assessment using PCA and LDA

Abstract

It is possible to identify voice disorders using certain features of speech signals. A complementary technique could be acoustic analysis of the speech signal, which is shown to be a potentially useful tool to detect voice diseases. The focus of this study is to formulate a speech parameter estimation algorithm for analysis and detection of vocal fold pathology and also bring out scale to measure severity of the disease. The speech processing algorithm proposed estimates features necessary to formulate a stochastic model to characterize healthy and pathology conditions from speech recordings. Speech signal features such as MFCC are extracted from acoustic analysis of voiced speech of normal and pathological subjects. A principal component analysis with minimum distance classifier (PCA+MDC) and linear discriminant analysis (LDA) classifier are designed and the classification results have been reported.