Gender–dependent GMM–UBM for tracking Parkinson’s disease progression from speech

Konferenz: Speech Communication - 12. ITG-Fachtagung Sprachkommunikation
05.10.2016 - 07.10.2016 in Paderborn, Deutschland

Tagungsband: ITG-Fb. 267: Speech Communication

Seiten: 5Sprache: EnglischTyp: PDF

Persönliche VDE-Mitglieder erhalten auf diesen Artikel 10% Rabatt

Autoren:
Arias-Vergara, Tomas; Vasquez-Correa, Juan Camilo; Vargas-Bonilla, Jesus Francisco (Faculty of Engineering, Universidad de Antioquia UdeA, Calle 70 No. 52-21, Medellin, Colombia)
Orozco-Arroyave, Juan Rafael (Faculty of Engineering, Universidad de Antioquia UdeA, Calle 70 No. 52-21, Medellin, Colombia & Pattern Recognition Lab, Friedrich-Alexander-Universität Erlangen-Nürnberg, Germany)
Haderlein, Tino; Noeth, Elmar (Pattern Recognition Lab, Friedrich-Alexander-Universität Erlangen-Nürnberg, Germany)

Inhalt:
Parkinson’s disease (PD) severity is evaluated by neurologist experts by means of several tests. One of them is the Movement Disorder Society-Unified Parkinson’s Disease Rating Scale (MDS-UPDRS). The main hypothesis is that changes in the speech of PD patients reflect changes in their neurological state. In this study we use the Gaussian Mixture Model-Universal Background Model approach to track the disease progression per speaker. Speech recordings from 62 PD patients were captured from 2012 to 2015 in three recording sessions. The validation of the models is performed with recordings of 7 patients (3 male and 4 female). The models were trained using speech recordings from male and female patients separately. According to the results, it is possible to track the disease progression with a Pearson’s correlation of up to 0:88 for males and 0:53 for females.