Researchers have built a "speech clock" that estimates a person's chronological age from hundreds of acoustic and linguistic features of how they talk, according to a study published in Science Advances. The gap between a person's real age and the age predicted from their speech - what the team calls the speech age gap - tracked with several independent markers of biological aging, brain health, cognition, social adversity and dementia.
The analysis covered 2,928 Spanish-speaking participants from Argentina, Chile, Colombia, Mexico and Peru, a group that included healthy adults alongside people with mild cognitive impairment, Alzheimer's disease and different forms of frontotemporal dementia. Instead of isolating one vocal property, the models pulled together speech rate and pauses, pitch, emotional content, vocabulary, semantic precision, and the volume and organisation of what people actually said.
People whose speech sounded older than their birth certificate suggested showed signs of accelerated aging in other systems too. The speech age gap lined up with brain age derived from structural and functional neuroimaging, and with epigenetic aging measured by three separate DNA-methylation clocks, which estimate biological age from age-related chemical changes in DNA.
That convergence is the interesting part. Brain imaging and methylation assays are expensive and slow; a recording of someone speaking is neither. If the speech age gap holds up as a proxy for those harder-to-get measures, it could give researchers a cheap, repeatable way to track aging and cognitive decline - particularly valuable in regions where neuroimaging and lab-based biomarker testing are thin on the ground. The cohort here was entirely Spanish-speaking and drawn from five Latin American countries, which makes the work unusual in a field dominated by English-language and North American or European samples.