Speech-Based Aging Clock Shows Promise for Dementia Assessment

A new study used speech recordings from 2,928 participants in five Latin American countries to build an aging clock based on acoustic and linguistic features. The resulting speech age gaps separated healthy controls from people with mild cognitive impairment, Alzheimer's disease, and frontotemporal dementia, and they tracked with tau levels, brain measures, and epigenetic age. The authors suggest this approach could become a scalable, low-cost biomarker for aging and dementia.

EXPANDED:

Supervised models used acoustic and linguistic speech features to estimate chronological age. The gap between estimated and actual age served as a marker; positive values meant older-sounding speech. This gap separated healthy controls from patient groups, with Alzheimer's disease below non-language-dominant frontotemporal dementia,