| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 259.27 KB | Adobe PDF |
Advisor(s)
Abstract(s)
Este trabalho tem como objetivo diferenciar distúrbios vocais relacionados com nódulo vocal e edema de Reinke através de parâmetros não-lineares. Os parâmetros são obtidos da banda completa e, por meio da transformada wavelet, de sub-bandas de vogal sustentada. A wavelet que maximiza a capacidade discriminante individual dos parâmetros é buscada. Duas seleções dos parâmetros em sub-bandas são realizadas. Os parâmetros em banda completa e em sub-bandas, e os conjuntos selecionados são aplicados a vários classificadores com validação cruzada leave-one-out. Acurácias de classificação de até 86,2% são obtidas sem seleção de parâmetros enquanto acurácias até 100% são alcançadas com parâmetros selecionados.
This work aims to detect vocal disorders related to vocal nodule and Reinke’s edema through non-linear features. The features are computed from the fullband and, by means of the wavelet transform, from sub-bands of sustained vowel. The wavelet that maximizes the individual discriminating capacity of the features is sought. Two sub-band parameter selections are performed. The fullband and sub-band parameters, and the selected sets are applied to several classifiers with leave-one-out cross-validation. Classification accuracies of up to 86.2% are obtained without parameter selection while accuracies of up to 100% are achieved with selected parameters.
This work aims to detect vocal disorders related to vocal nodule and Reinke’s edema through non-linear features. The features are computed from the fullband and, by means of the wavelet transform, from sub-bands of sustained vowel. The wavelet that maximizes the individual discriminating capacity of the features is sought. Two sub-band parameter selections are performed. The fullband and sub-band parameters, and the selected sets are applied to several classifiers with leave-one-out cross-validation. Classification accuracies of up to 86.2% are obtained without parameter selection while accuracies of up to 100% are achieved with selected parameters.
Description
Keywords
Nódulo vocal Edema de Reinke Parâmetros não-lineares Wavelet Classificação Vocal nodule Reinke’s edema Nonlinear features Classification