Speech biomarkers
for clinical trials
A speech analysis framework for pharmaceutical and clinical research partners. Deep-learning embeddings, acoustic-linguistic and cognitive-linguistic features, combined into explainable digital measures.
Why speech
A short speech sample carries acoustic, linguistic, and prosodic information that maps onto the cognitive and motor domains pharma trials measure.
Short structured tasks completed remotely
No clinician time for home acquisition. Suitable for older adults and decentralised trial designs.
Trajectory information between clinic visits
Frequent acquisition between visits supports longitudinal trajectory analysis.
Every reading traces back to its features
Each reading traces to the acoustic and linguistic features that drove it, for study teams and medical review.
How it works
Our pipeline
One recording, three streams, one multi-stream model, read on the clinical scale.
What we measure
The scales each indication runs, read from speech.
Mobile device audio processing
How readings hold when a device’s audio processing changes.
Therapeutic areas
Six programs, each mapped to the endpoints its trials already run.
Alzheimer's disease
Readings on the MMSE and MoCA, followed between visits.
Parkinson's disease
Readings on the MDS-UPDRS speech item, followed between visits.
ALS
Readings on the ALSFRS-R speech item, followed between visits.
Huntington's disease
Speech and language change from the premanifest stage, mapped to the cUHDRS spoken components.
DM1 and neuromuscular
Articulation, phonation and respiratory support, mapped to the endpoints these trials run.
Depression and BPSD
A second reading of the recorded rating interview.
Where would speech sit in your next protocol?
Share your therapeutic area and current studies.
Get in touch →