An automated computer tool called a “speech clock” can estimate a person’s biological ageing by analysing how they speak, according to a new study. Researchers found that a voice contains data reflecting what is happening inside the body, offering a potential low-cost, non-invasive alternative for measuring biological ageing across cognition, the brain, and systemic biology.
How the Speech Clock Evaluates Biological Ageing
The automated tool calculates a person’s speech age and compares it against their actual birth age. This process reveals whether an individual sounds younger, average, or older than expected chronologically.
“Our voice appears to contain much more information about aging than we previously recognised,” said Agustín Ibáñez, senior author of the study at Trinity College Dublin. He noted that the method captures the passage of time alongside signals from cognition, systemic biology, and accumulated social environments.
To build the tool, researchers analysed audio recordings from nearly 3,000 Spanish-speaking adults across Latin America. Participants completed multiple tasks, including narrating an animated short film, retelling a story after a 20-minute gap, and naming items within specific categories.
Machine learning examined hundreds of features within the recordings. These included speech rate, pauses, pitch, emotional content, vocabulary, and semantic precision.
Speech Patterns Linked to Brain Structure and Alzheimer’s Protein
The speech clock successfully distinguished healthy participants from individuals diagnosed with dementia. Healthy individuals displayed the lowest speech age gaps, while progressively wider gaps appeared in participants with Alzheimer’s disease and various forms of frontotemporal dementia.
Participants whose speech sounded older than their chronological age exhibited accelerated ageing signs across several biological systems. MRI brain scans revealed greater loss of brain structure and altered brain activity in this group.
Older speech scores also correlated with higher levels of p-Tau217, a blood protein tied directly to Alzheimer’s disease. These participants scored lower on tests measuring memory, focus, and problem-solving, affecting both language tasks and non-verbal tests like visual memory.
“The broader finding is nevertheless striking in that a person’s voice may provide a remarkably compact readout of multiple dimensions of aging,” Ibáñez added.
Traditional Medical Assessments Versus Remote Audio Testing
Conventional measures for chronological age, brain ageing, cognition, and Alzheimer’s phenotypes typically rely on expensive tools such as MRI scanners, blood draws, and specialised clinical assessments. In contrast, the speech clock functions remotely on standard devices.
The method requires no specialist medical staff to administer the voice tasks. However, the study authors emphasised that the speech clock does not yet function as a diagnostic test for dementia.
The current research model cannot determine whether an older-appearing speech profile predicts which individuals will subsequently develop cognitive decline.
Frequently Asked Questions About the Speech Clock Study
How many people participated in the speech clock research?
The study analysed audio recordings from nearly 3,000 Spanish-speaking adults across Latin America.
What specific tasks did study participants perform?
Participants narrated a short animated film, retold a story after a 20-minute delay, and named items belonging to given categories.
Related reading