A vocal AI clock developed at Trinity College Dublin can estimate the speed of aging based on a few minutes of recorded speech, according to a study published on September 30, 2026, in the journal Science Advances. International researchers tested this idea by applying artificial intelligence to the voices of 2,928 adults aged 18 to 88 years, followed in Latin America. The study links vocal patterns to brain aging, body aging, and dementia risk. At the heart of the device is a "vocal clock," or "speech clock," which estimates chronological age from hundreds of acoustic and linguistic characteristics, then calculates a "gap between real age and predicted age." "Our voice seems to contain much more information about aging than we thought," explains Agustin Ibanez, Professor of Neurology at the Global Brain Health Institute and at the School of Medicine of Trinity College Dublin. The researchers, led by Hernan Hernandez, analyzed standardized Spanish recordings from five Latin American countries. The algorithm considers more than just the timbre of the voice. It takes into account speech rate and pauses, pitch, emotional expression, vocabulary, semantic precision, and the quantity and organization of speech. These combined indicators allow the estimation of a vocal age without blood tests or MRI scans. The model first predicts real age and then calculates for each person a vocal gap. When this gap is positive, the speech appears "older" than the civil status. In the study, these gaps were associated with cerebral aging measured by structural and functional neuroimaging, with epigenetic aging based on DNA methylation, and with more fragile cognitive health.