Exploring Speech Subtypes in Individuals with Parkinson’s Disease based on Acoustic, Aerodynamic, and Perceptual Features
Communication Sciences & Disorders · 30.06.2026 · 31 (2) · 464-476
DOI: 10.12963/csd.260222 ↗Kurz gefasst
Fachartikel in der Zeitschrift Communication Sciences & Disorders, online erschienen am 30.06.2026. Verfasst von HeeBaek Shin. Die Arbeit nennt 24 Literaturangaben.
Abstract (Originalsprache)
Objectives
This study aimed to explore speech subtypes in individuals with Parkinson’s disease (PD) based on acoustic, aerodynamic, and perceptual features and to identify whether the derived clusters could be interpreted as distinct speech subtypes.
Methods
Forty-four individuals with PD participated in this study. Five acoustic/aerodynamic variables (pitch range, sound pressure level, mean airflow rate, speaking rate, and articulation rate) and four perceptual variables (overall speech intelligibility, loudness, speaking rate, and articulation accuracy) were included in the final cluster analysis. Perceptual ratings were obtained from five listeners using visual analog scales. Hierarchical clustering and kmeans clustering were performed based on standardized z-scores. Group differences were examined using one-way ANOVA, and effect sizes (η2) were calculated to determine the relative contribution of each variable. Listener-perceived speech subtype frequencies and background variables were also compared across clusters.
Results
A three-cluster solution was identified. The clusters were labeled as Low loudness–slow rate type (LL-SR, n=14), Mild speech inefficiency type (MSI, n=24), and Fast rate–articulatory deficit type (FR-AD, n=6). Significant group differences were found across all nine variables. The variables with the largest effect sizes were overall speech intelligibility, articulation accuracy, loudness, and sound pressure level. Listener-perceived frequency analysis showed significant group differences in articulatory deficit frequency and mixed-group frequency. Among background variables, only sex differed significantly across clusters.
Conclusion
Speech characteristics in PD appear to be heterogeneous and can be organized into three distinguishable subtypes. The final cluster structure was primarily differentiated along the dimensions of intelligibility, articulation, and loudness, suggesting the need for more individualized assessment and intervention approaches.
Zitieren
APA
Shin, H. (2026). Exploring Speech Subtypes in Individuals with Parkinson’s Disease based on Acoustic, Aerodynamic, and Perceptual Features. Communication Sciences & Disorders, 31(2), 464-476. https://doi.org/10.12963/csd.260222
Vancouver
Shin H. Exploring Speech Subtypes in Individuals with Parkinson’s Disease based on Acoustic, Aerodynamic, and Perceptual Features. Communication Sciences & Disorders. 2026;31(2):464-476. doi:10.12963/csd.260222
BibTeX
@article{Shin2026,
author = {Shin, HeeBaek},
title = {Exploring Speech Subtypes in Individuals with Parkinson’s Disease based on Acoustic, Aerodynamic, and Perceptual Features},
journal = {Communication Sciences & Disorders},
year = {2026},
volume = {31},
number = {2},
pages = {464--476},
doi = {10.12963/csd.260222}
}
EUVIMED