Perceptual Analysis of Speech Signals from People with Parkinson's Disease

Parkinson’s disease (PD) is a neurodegenerative disorder of the nervous central system and it affects the limbs motor control and the communication skills of the patients. The evolution of the disease can get to the point of affecting the intelligibility of the patient’s speech. The treatments of the PD are mainly focused on improving limb symptoms and their impact on speech production is still unclear. Considering the impact of the PD in the intelligibility of the patients, this paper explores the discrimination capability of different perceptual features in the task of automatic classification of speech signals from people with Parkinson’s disease (PPD) and healthy controls (HC). The experiments presented in this paper are performed considering the five Spanish vowels uttered by 20 PPD and 20 HC. The considered set of features includes linear prediction coefficients (LPC), linear prediction cepstral Coefficients (LPCC), Mel-frequency cepstral coefficients (MFCC), perceptual linear prediction coefficients (PLP) and two versions of the relative spectra coefficients (RASTA). Accordin the results for vowels /e/ and /o/ it is not enough to consider one kind of perceptual features, it is required to perform combination of different coefficients such as PLP, MFCC and RASTA. For the case of the remaining vowels, the best results are obtained considering only one kind of perceptual features, PLP for vowel /a/ and MFCC for vowels /i/