Phonological posteriors and GRU recurrent units to assess speech impairments of patients with Parkinson’s disease

Vasquez Correa J, Garcia N, Orozco-Arroyave JR, Cernak M, Nöth E (2018)

Publication Language: English

Publication Type: Book chapter / Article in edited volumes

Publication year: 2018

Publisher: Springer Interational

Edited Volumes: Text, Speech, and Dialogue

Series: Lecture Notes in Computer Science

Book Volume: 11107

Pages Range: 453-461

ISBN: 978-3-030-00794-2

URI: https://link.springer.com/chapter/10.1007/978-3-030-00794-2_49

DOI: 10.1007/978-3-030-00794-2_49

Abstract

Parkinson’s disease is a neurodegenerative disorder characterized by a variety of motor symptoms, including several impairments in the speech production process. Recent studies show that deep learning models are highly accurate to assess the speech deficits of the patients; however most of the architectures consider static features computed from a complete utterance. Such an approach is not suitable to model the dynamics of the speech signal when the patients pronounce different sounds. Phonological features can be used to characterize the voice quality of the speech, which is highly impaired in patients suffering from Parkinson’s disease. This study proposes a deep architecture based on recurrent neural networks with gated recurrent units combined with phonological posteriors to assess the speech deficits of Parkinson’s patients. The aim is to model the time-dependence of consecutive phonological posteriors, which follow the sound patterns of English phonological model. The results show that the proposed approach is more accurate than a baseline based on standard acoustic features to assess the speech deficits of the patients.

Authors with CRIS profile

Juan Vasquez Correa Professur für Informatik (Mustererkennung) Elmar Nöth Professur für Informatik (Mustererkennung)

Involved external institutions

Universidad de Antioquía (UDEA) CO Colombia (CO) idiap Research Institute CH Switzerland (CH)

How to cite

APA:

Vasquez Correa, J., Garcia, N., Orozco-Arroyave, J.R., Cernak, M., & Nöth, E. (2018). Phonological posteriors and GRU recurrent units to assess speech impairments of patients with Parkinson’s disease. In Prof. Dr. Petr Sojka, Aleš Horák, Ivan Kopeček, Karel Pala (Eds.), Text, Speech, and Dialogue. (pp. 453-461). Springer Interational.

MLA:

Vasquez Correa, Juan, et al. "Phonological posteriors and GRU recurrent units to assess speech impairments of patients with Parkinson’s disease." Text, Speech, and Dialogue. Ed. Prof. Dr. Petr Sojka, Aleš Horák, Ivan Kopeček, Karel Pala, Springer Interational, 2018. 453-461.

BibTeX: Download