Your Voice Could Reveal How Well You’re Aging, New Research Suggests - Video

A machine-learning “speech clock” may estimate chronological age and provide clues about brain ageing and cognition, according to a study published in Science Advances.

Researchers analysed 2,928 Spanish-speaking adults from Argentina, Chile, Colombia, Mexico and Peru. Participants included healthy adults and people with mild cognitive impairment, Alzheimer’s disease and frontotemporal dementia.

The researchers used machine learning to analyse hundreds of speech features, including speaking speed, pauses, pitch, vocabulary, meaning and speech organisation. These features were combined to estimate chronological age and calculate a “speech age gap”, showing the difference between actual and speech-predicted age.

A larger speech age gap was associated with markers of accelerated ageing. It was linked to brain age measured through neuroimaging and higher epigenetic age assessed using DNA-methylation clocks.

The speech age gap was also associated with poorer overall cognition, executive function, daily functioning and memory. The associations also extended to non-language cognitive tests, suggesting broader brain effects.

The speech-based measure also differed between clinical groups. Healthy participants had the smallest speech age gaps, while larger gaps were observed among people with Alzheimer’s disease and frontotemporal dementia. In Alzheimer’s disease, the speech measure was also associated with higher plasma p-tau217, a blood biomarker linked to Alzheimer’s pathology.

Speech ageing was additionally associated with greater social disadvantage, including differences in education, financial conditions, food security, healthcare access and early-life experiences.

However, the findings do not establish that an older-appearing speech pattern predicts future dementia. The study was primarily cross-sectional, meaning it cannot determine cause and effect or establish whether speech changes precede cognitive decline.

Researchers said longitudinal studies, validation in additional languages and testing in natural speech settings are needed.

If confirmed, speech-based tools could eventually provide a low-cost, non-invasive method for monitoring ageing and brain health.

REFERENCE: Agustin Ibanez, et al.; Your voice may reveal how fast and how well you’re aging; Science Advances; DOI: 10.1126/sciadv.aef9864