Linear prediction and discrete wavelet transform to identify pathology in voice signals

Nenhuma Miniatura disponível

Data

2017-09-28

Autores

Fonseca, Everthon Silva [UNESP]
Pereira, Denis Cesar Mosconi
Maschi, Luis Fernando Castilho
Guido, Rodrigo Capobianco [UNESP]
Paulo, Katia Cristina Silva [UNESP]

Título da Revista

ISSN da Revista

Título de Volume

Editor

Resumo

This work describes an algorithm to help in the identification of pathologically affected voices. Based on inverse linear prediction filter (LPC) and discrete wavelet transform (DWT), this method can be used in conjunction with other classifiers in order to improve them, by the addition of the new parameter we propose, DWT-RMS. Using no association with other methods, DWT-RMS gives quantitative evaluation of voice signals from male and female subjects of different ages and leads to an adequate larynx pathology classifier with 85.94% of classification accuracy, 0% of false negatives and 14.06% of false positives, to identify nodules in vocal folds.

Descrição

Palavras-chave

pathologies, prediction, signals, voice, wavelet

Como citar

2017 Signal Processing Symposium, SPSympo 2017.