The researchers conducted a comprehensive investigation of the reliability of the speech task battery across seven cognitive tasks and about one hundred speech acoustic features. This study provides crucial data support for feature engineering.
To address the methodological challenge of assessing the reliability of online tests, they also refined a split-half reliability estimation method. The clinical efficacy of the AI-based speech assessment system was validated using a valuable sample of individuals with Wilson Disease, which served as an ideal disease model with mixed dysarthria phenotypes.
The researchers demonstrated the superior performance of their model in predicting dysarthria severity. Their systematic methodological framework provided valuable insights into AI-driven intelligent diagnosis, rehabilitation, and home support for