Acoustic and kinematic predictors of intelligibility and articulatory precision in Parkinson’s disease

Abstract

Purpose: This study investigated relationships within and between perceptual, acoustic, and kinematic measures in speakers with and without dysarthria due to Parkinson’s disease (PD) across different clarity conditions. Additionally, the study assessed the predictive capabilities of selected acoustic and kinematic measures for intelligibility and articulatory precision ratings.Methods: Forty participants, comprising 22 with PD and 18 controls, read three phrases aloud under conversational, less clear, and more clear speaking conditions. Acoustic measures and their theoretical kinematic parallel measures (i.e., acoustic and kinematic distance and vowel space area [VSA]; F2 slope and kinematic speed) were obtained from the diphthong /aɪ/ and selected vowels in the sentences. A total of 368 listeners from crowdsourcing provided ratings for intelligibility and articulatory precision. The research questions were examined using correlations and linear mixed-effects models.Results: Overall, intelligibility and articulatory precision ratings were highly correlated across all speakers. Acoustic and kinematic distance showed a moderate correlation, while acoustic and kinematic VSA exhibited no correlation. Among the measures, acoustic VSA and kinematic distance were robust predictors of both intelligibility and articulatory precision ratings.Conclusions: The positive predictive relationships observed between acoustic VSA and kinematic distance in relation to intelligibility and articulatory precision ratings support the use of behavioral modifications aimed at eliciting larger articulatory gestures to improve intelligibility in individuals with dysarthria due to PD.

The Effects of Electromagnetic Articulography Sensors on Speech in Individuals With and Without Dysarthria

“…However, relying solely on acoustic signals to infer articulatory gestures is complicated by the fact that multiple articulatory gestures can produce the same acoustic signal (Brunner & Hoole, 2012;Hughes & Abbs, 1976;Perkell et al, 1993;Perrier & Fuchs, 2015). Thus, kinematic analyses using EMA to quantify articulatory working space (Lee et al, 2017;Thompson & Kim, 2024a, 2024b, tongue displacement and speed (Mefferd, 2015;Rong et al, 2012), and movement variability (Chu et al, 2020) are particularly beneficial for studying speakers with and without speech disorders, such as dysarthria.…”