SPEECH CLOCK TOOL LINKS VOICE PATTERNS TO BRAIN AGEING

Scientists have developed a computer tool known as a "speech clock" that estimates a person's age by analysing how long they pause, their pitch and their vocabulary, according to a study published in the journal Science Advances. The tool found that younger-sounding speech tends to be faster and more fluent with shorter pauses, while older-sounding speech shows the opposite pattern. Researchers tested the tool on 2,928 Spanish-speaking participants from Argentina, Chile, Colombia, Mexico and Peru, including healthy adults and people with mild cognitive impairment, Alzheimer's disease and different forms of frontotemporal dementia. In some cases, the tool judged a participant's voice to sound older than their actual age. Those participants were more likely to show signs of biological ageing, cognitive decline and dementia, the study found.

Researchers calculated what they called a "speech age gap", the difference between a participant's chronological age and the biological age suggested by their voice. Participants whose speech age exceeded their chronological age tended to perform worse on memory, thinking and planning tests, even when those tests did not involve speaking. Larger speech age gaps were also linked to older-looking brain scans and DNA changes associated with ageing, according to the study. Agustin Ibanez, professor in Brain Health at the Global Brain Health Institute and School of Medicine, Trinity, and senior author of the study, said voices appear to contain more information about ageing than previously recognised.

Ibanez said the findings suggest speech captures chronological time alongside signals from cognition, the brain, systemic biology and accumulated social environment. He said speech clocks, potentially combined with other biomarkers, could eventually complement more expensive measures of ageing. The findings point towards voice analysis as a possible screening method to identify patients at risk of dementia and cognitive decline. The researchers did not set out a timeline for when such a tool might be used in clinical settings.