Features based on quasi-sinudoidal modeling for tremor detection in Parkinsonian voice

Author

Ben Rhouma, Alaa ; Ben Jebara, Sofia

Author_Institution

COSIM Lab., Carthage Univ., Ariana, Tunisia

fYear

2014

fDate

17-19 March 2014

Firstpage

434

Lastpage

439

Abstract

This paper aims defining features to characterize Parkinsonian voice affected by tremor. It uses quasi-sinusoidal modelling of signals which assumes that speech signal is a sum of sinusoids with time-linearly varying instantaneous amplitudes and frequencies permits. The parameters of this model are calculated and their behavior is analyzed. The statistical analysis using box-plots permits to show the ability of this model to discriminate the Parkinsonian voice from the healthy voice.

Keywords

acoustic signal detection; acoustic signal processing; diseases; feature extraction; medical signal processing; neurophysiology; physiological models; signal classification; speech processing; statistical analysis; waveform analysis; Parkinsonian voice characterization; Parkinsonian voice classification; Parkinsonian voice features; Parkinsonian voice tremor detection; box-plots; model parameter behavior analysis; model parameter calculation; quasi-sinudoidal modeling based features; sinusoid sum; speech signal model; statistical analysis; time-linear instantaneous amplitude variation; time-linear instantaneous frequency variation; Databases; Harmonic analysis; Noise; Parkinson´s disease; Speech; Time-frequency analysis; Parkinson´s disease; features for discrimination; quasi-sinusoidal model; tremor;

fLanguage

English

Publisher

ieee

Conference_Titel

Advanced Technologies for Signal and Image Processing (ATSIP), 2014 1st International Conference on

Conference_Location

Sousse

Type

conf

DOI

10.1109/ATSIP.2014.6834651

Filename

6834651