AI-driven speech biomarkers for
motor
function
ki:e SB-M
SB-M by ki:elements
objective, scalable
speech-based motor assessments
QUANTIFY MOTOR SYMPTOMS THROUGH EVERYDAY SPEECH
SB-M
SB-M is an AI-driven digital outcome assessment for motor impairments in Parkinson’s disease (PD), amyotrophic lateral sclerosis (ALS), multiple system atrophy (MSA), and related disorders. It enables sensitive, reliable, and remote-capable monitoring of motor function, enhancing clinical trials from early detection to disease progression tracking. 2 gold standard speech tasks (Sustained phonation and PaTaKa), a reading task, and a monologue task are performed in less than 5 minutes, and 100 features are automatically extracted.
clinically meaningful
aspect of communicationAffected individuals value and seek to preserve intelligibility, and it can be measured objectively & automatically through conversational speech.
Explore further into the following disease areas we support:
Built for real-world clinical research
OBJECTIVE AND AUTOMATED
Remove subjectivity and variability with consistent, high-quality scoring—anytime, anywhere.
Sensitive to Speech Subsystems
Measure changes in articulation, fluency, and prosody with precision for a deeper view of disease impact.
motor-fluctuation resilient
Accurately track ON/OFF states in individuals with Parkinson’s.
PATIENT-CENTRIC and MEANINGFUL
Quantify what matters with our Intelligibility Score.
Transparent and Trackable
View assessment progress and completion to support consistent longitudinal monitoring.
Effortlessly Integrated
Fit into existing workflows without added burden on sites or patients. Use it out-of-the-box with our Mili System or integrate into third party systems.
Smarter
Algorithms —Stronger
EvidenceValidation based on global cohorts following approved methodology from the V3 framework of the Digital Medicine Society (DiME)
SB-M reliably differentiates between motor disorder phenotypes (PD, ALS, HD, PSP) and healthy controls, demonstrating clinical validity across diseases, languages, and sites.
Automatic speech analysis combined with machine learning reliably predicts motor state in Parkinson's disease
Thies et al. (2025) | doi: 10.1038/s41531-025-00959-4
An automatic measure for speech intelligibility in dysarthrias-validation across multiple langagues and neurological disorders
Tröger et al. (2024) | doi: 10.3389/fdgth.2024.1440986
your
study?