ki:elements

The ki: SB-M intelligibility score — An automatic measure for intelligibility in motor speech disorders

Johannes Tröger, Felix Dörr, Louisa Schwed, Nicklas Linz, Alexandra König, Tabea Thies, Michael T. Barbe, Juan Rafael Orozco-Arroyave & Jan Rusz

* Poster presented at the International Congress of Parkinson’s Disease and Movement Disorders, Philadelphia (USA)

Abstract

Objective: To validate the ki: SB-M intelligibility score, a digital measure for speech intelligibility, across various motor speech disorders and languages in accordance with the Digital Medicine Society (DiMe) V3 framework.

Background: Dysarthria, a motor speech disorder resulting from muscle weakness or paralysis, significantly impairs speech intelligibility and quality of life. It is prevalent in conditions such as Parkinson’s Disease (PD), atypical parkinsonism (e.g., progressive supranuclear palsy [PSP]), Huntington’s Disease (HD), and amyotrophic lateral sclerosis (ALS). Enhancing and measuring intelligibility of speech is crucial for patients and serves as an important endpoint in clinical research and drug development.

Methods: The study analyzed four datasets comprising healthy controls (HC) and patients with PD, HD, PSP, and ALS from Czech, Colombian, and German populations. Speech intelligibility was measured using the ki: SB-M intelligibility score, derived from automatic speech recognition (ASR) systems. Verification involved assessing inter-ASR reliability and temporal consistency. Analytical validation was conducted through correlations with gold standard clinical dysarthria scores for each disease. Clinical validation included group comparisons between HC and patients.

Results: Verification indicated good to excellent inter-rater reliability between ASR systems and fair to good temporal consistency. Analytical validation showed significant correlations between the SB-M intelligibility score and established clinical measures of speech impairment across all patient groups and languages. Clinical validation revealed significant differences in intelligibility scores between pathological groups and healthy controls, demonstrating the measure’s discriminative capability.

Conclusion: The ki: SB-M intelligibility score is a reliable, valid, and clinically relevant tool for assessing speech intelligibility in motor speech disorders. It offers potential for enhancing clinical trials through automated, objective, and scalable assessments. Future research should investigate its utility in monitoring disease progression and therapeutic efficacy, as well as expand validation to include additional dysarthria cases.

Share this article