Trinity College Dublin researchers develop speech clock tool

A novel automated computer tool called a “speech clock” can estimate a person’s chronological age by analyzing hundreds of acoustic and linguistic characteristics from a simple voice recording. Developed by researchers at Trinity College Dublin and international partners, this non-invasive approach links older-sounding speech patterns to neurodegeneration, cognitive decline, and biological ageing markers.

In Plain English: The Clinical Takeaway

  • What a speech clock does: It uses machine learning to listen to audio recordings, analyzing variables like speech rate, pitch, pauses, and vocabulary to calculate a person’s estimated “speech age.”
  • Why the gap matters: When a person’s speech age is older than their actual birth age, it correlates with accelerated biological ageing and structural brain changes observed on scans.
  • Current limitations: Agustín Ibáñez pointed out that this system cannot yet serve as a diagnostic tool for dementia, nor can it determine if an advanced speech age profile forecasts future cognitive decline.

Evaluating the Speech Clock in Multinational Cohorts

Published in the journal Science Advances, the study analyzed audio recordings from 2,928 Spanish-speaking adults across Argentina, Chile, Colombia, Mexico, and Peru. Participants ranged in age from 18 to 88 years old, encompassing both healthy individuals and patients diagnosed with mild cognitive impairment, Alzheimer’s disease, and various forms of frontotemporal dementia. The research team utilized data from the ReD-Lat consortium to evaluate these diverse populations.

To capture comprehensive linguistic and acoustic data, participants performed a variety of verbal tasks. These tasks involved recounting a brief animated movie, reiterating a narrative following a 20-minute interval, and identifying objects belonging to specified groups. Machine-learning algorithms processed over 700 distinct features from these recordings. The analytical framework measured speaking velocity, frequency of pauses, vocal pitch, emotional tone, and semantic precision to generate an individualized “speech age gap”—the mathematical difference between chronological age and the age predicted by vocal biomarkers.

Correlating Vocal Profiles with Brain Structure and Blood Biomarkers

The clinical implications of a wide speech age gap extend into systemic biology and neuroanatomy. Individuals exhibiting older-sounding speech profiles demonstrated distinct markers of accelerated ageing across multiple physiological systems. Structural and functional neuroimaging revealed that larger speech age gaps corresponded with brain age. Furthermore, biological validation came from molecular assays of blood plasma. Participants with Alzheimer’s disease who displayed older speech scores also presented elevated levels of p-Tau217. This phosphorylated tau protein serves as a blood-based biomarker for Alzheimer’s pathology. Additionally, epigenetic clocks derived from DNA-methylation analysis confirmed that older speech profiles tracked alongside chemical modifications in DNA traditionally associated with cellular ageing.

Comparative Metrics of the Speech Clock Study
Biological or Clinical System Measurement Tool / Assay Observed Association with Older Speech Scores
Brain Structure & Function Structural and Functional Neuroimaging Association with brain age
Molecular Pathology Blood Plasma Biomarkers Elevated levels of p-Tau217 linked to Alzheimer’s disease
Epigenetic Ageing DNA-Methylation Clocks Older biological age readings across independent DNA assays
Cognitive Performance Neuropsychological Testing Poorer global cognition, executive function, functional abilities, and memory

Socioeconomic Influences and Diagnostic Accessibility

The research uncovered intersections between vocal biomarkers and social determinants of health. Participants whose speech appeared older than their chronological age frequently encountered socioeconomic disadvantages, including lower education, financial hardship, and food insecurity. Agustín Ibáñez noted that the voice appears to capture a cumulative readout spanning chronological time, cognitive status, systemic biology, and accumulated environmental stressors.

Trinity College Dublin researchers develop speech clock tool
Photo: the-independent.com

Despite these broad correlations, clinical experts urge caution regarding the tool’s current clinical utility. Dr. Manisha Parulekar, co-director of the Center for Memory Loss and Brain Health at Hackensack University Medical Center, pointed out that vocal output reflects a person’s general physical and mental state. A person’s speech may sound older simply because they are severely depressed, exhausted, or navigating serious life stress.

Contraindications & When to Consult a Doctor

Because the speech clock is currently an investigational research tool rather than a validated diagnostic instrument, it cannot be used at home to self-diagnose dementia, Alzheimer’s disease, or accelerated brain ageing. Patients and caregivers must recognize that vocal changes can stem from non-neurological etiologies.

An older woman with red hair sits behind a desk and talks on the phone
Photo: Live Science

Individuals experiencing persistent cognitive deficits should consult a primary care physician or a board-certified neurologist. Formal clinical evaluation requires comprehensive neurocognitive testing, validated laboratory assays, and neuroimaging rather than acoustic screening alone.

Future Directions in Remote Health Screening

Traditional assessments of biological ageing often demand resource-intensive interventions, including MRI scanners, blood samples, molecular assays, or specialized clinical assessments. In contrast, the speech clock framework operates via standard recording devices without requiring specialist medical staff or physical travel. Researchers emphasize that before clinical implementation can occur, longitudinal studies, cross-cultural validation in additional languages, and testing in naturalistic acoustic environments remain necessary.

Trinity College Dublin researchers develop speech clock tool
Photo: News-Medical

  • Science Advances: Machine-learning speech clocks estimate chronological age and reflect brain ageing across multinational cohorts.

Seymour Hersh Iran speech at Trinity College Dublin part 1
EU Researchers' Night – Discover Research Trinity College Dublin
Photo of author

Priya Deshmukh - Senior Editor, Health

Priya Deshmukh Senior Editor, Health Deshmukh is a practicing physician and renowned medical journalist, honored for her investigative reporting on public health. She is dedicated to delivering accurate, evidence-based coverage on health, wellness, and medical innovations.

Western Bulldogs Defeat St Kilda in Low-Scoring AFLW Clash