TL;DR
Researchers used hundreds of speech features to estimate chronological age in 2,928 Spanish-speaking adults across five Latin American countries. The gap between predicted and actual age was associated with measures of brain and biological aging, cognition, social adversity and dementia, but the study does not show that the tool can predict future decline or diagnose dementia.
Researchers have developed a machine-learning “speech clock” that estimates chronological age from a person’s speech, according to a study published in Science Advances. In 2,928 Spanish-speaking participants, the gap between predicted and actual age was associated with measures of brain and biological aging, cognitive performance, social adversity and dementia, although the researchers say the tool is not a diagnostic test.
Give the gift of listening to someone you love
- Choose a 1-, 3-, 6- or 12-month membership
- They pick the audiobooks they love
- Easy to send by email
The study included participants from Argentina, Chile, Colombia, Mexico and Peru. They included healthy adults and people diagnosed with mild cognitive impairment, Alzheimer’s disease or forms of frontotemporal dementia. The researchers used machine-learning models to combine hundreds of features, including speech rate and pauses, pitch, emotional content, vocabulary, semantic precision, and the amount and organization of verbal output.
The model generated an estimated age from speech. Researchers compared that estimate with each participant’s chronological age to calculate a speech age gap. Larger gaps were associated with brain age measured through structural and functional neuroimaging, and with epigenetic aging measured using three DNA-methylation clocks, according to the report.
Speech age gaps were also associated with poorer global cognition, executive function, functional abilities and several types of memory. The reported associations extended to non-linguistic cognitive measures as well as language tests. The study also found that healthy participants had the lowest gaps on average, with progressively larger gaps across Alzheimer’s disease and forms of frontotemporal dementia. The combined speech measure distinguished clinical groups better than individual speech features considered separately.
Speech links several aging measures
The findings suggest speech could offer a low-cost, non-invasive way to track multiple health signals, if later studies confirm the associations. Many measures used to study biological aging require imaging, blood samples, laboratory assays or specialist assessments. Speech can be recorded remotely and repeatedly, which could make it useful for research and monitoring in places with limited access to advanced diagnostic equipment.
The study’s inclusion of participants from five Latin American countries adds evidence from a region the report describes as historically underrepresented in dementia research. The authors say the results raise the possibility that speech measures, perhaps combined with other biomarkers, could complement more expensive approaches. That possibility remains a research prospect; this study does not establish that using the clock improves care or outcomes.
In Alzheimer’s disease, a larger speech age gap was associated with higher levels of plasma p-tau217, a blood biomarker related to Alzheimer’s pathology. The report also links accelerated speech aging with an adverse social exposome, including factors such as education, financial conditions, food insecurity, healthcare access and early-life experiences. These are associations; the findings do not show that any one factor causes a change in speech age.
Products Worth Considering

Hyperice Normatec 3 Leg Compression Boots, Standard Size

Coway Mighty Air Purifier AP-1512HH(W), White

Oral-B iO Series 7 Deep Clean + Protect Rechargeable Electric Toothbrush, Black, with 2 Brush Heads and Charging Travel Case
As an affiliate, we earn on qualifying purchases.
How the speech age gap works
The clock does not estimate age from a single voice trait. It combines acoustic features, such as pitch and pauses, with linguistic information about a person’s word choice, meaning and verbal output. The researchers then compare the model’s age estimate with the participant’s actual age. A person whose speech is estimated to be older has a larger positive gap, which the report describes as speech-age acceleration.
The research included people with different diagnoses, allowing the team to examine how the speech measure varied across clinical groups. It also compared the measure with imaging, DNA-methylation clocks, cognitive assessments and social factors. These comparisons show that the speech measure was related to several independent indicators in this sample. They do not establish that speech directly measures each underlying process.
The study’s design limits what can be concluded about time and cause. It was primarily cross-sectional, meaning participants’ speech and health measures were assessed in a way that cannot establish whether an older-appearing speech profile precedes later cognitive decline or dementia. The findings therefore concern observed relationships, not a forecast of an individual’s future health.
““Our voice appears to contain much more information about aging than we previously recognized.””
— Agustin Ibanez, senior author and professor in brain health at the Global Brain Health Institute and Trinity College Dublin
Prediction and diagnosis remain untested
The study does not show whether the speech clock can predict who will later develop cognitive decline or dementia. Its mainly cross-sectional design cannot establish that sequence. The researchers also say the measure is not currently a diagnostic test, so the findings do not support using it on its own to identify dementia in an individual.
It is also unclear how well the model would work with other languages and cultures, or with everyday speech recorded outside the study setting. The report does not provide a clinical threshold for interpreting an individual’s score. Further validation is needed before the measure could be considered for clinical use.
Long-term and multilingual studies needed
The researchers say longitudinal studies are needed to determine whether speech-age gaps track changes over time or precede later cognitive decline. Work in additional languages and cultures, along with testing in more naturalistic speech settings, would help establish how broadly the model applies.
Further studies could also assess how speech measures perform alongside existing biomarkers and clinical assessments. Until that evidence is available, the speech clock remains a research tool, and its associations should not be treated as an individual diagnosis or a prediction of future illness.
Key Questions
What is a speech clock?
It is a machine-learning model that estimates chronological age by combining acoustic and language features of speech, according to the study.
What does the speech age gap mean?
It is the difference between a participant’s actual age and the age estimated from speech. A larger gap means the model estimated an older age relative to the participant’s chronological age.
Can the speech clock diagnose dementia?
No. The researchers say it is not a diagnostic test. The study found associations between speech age gaps and clinical groups, but it did not establish a basis for diagnosing an individual.
Can it predict future cognitive decline?
That remains unknown. Because the study was primarily cross-sectional, it cannot show whether an older-appearing speech profile predicts later decline. Long-term studies are needed.
Who took part in the study?
The study analyzed 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 forms of frontotemporal dementia.
Source: rss
