started · updated
Trinity College Dublin researchers develop AI ‘speech clock’ to detect cognitive decline
Researchers at Trinity College Dublin have developed a machine learning model known as a ‘speech clock’ that can estimate a person’s biological age by analyzing vocal and linguistic patterns. The study, published in Science Advances, utilized data from 2,928 Spanish-speaking participants across Argentina, Chile, Colombia, Mexico, and Peru.
The model analyzes hundreds of features, including speech rate, pauses, pitch, vocabulary, semantic precision, and emotional content. By calculating a ‘speech age gap’—the difference between a person’s actual chronological age and their predicted speech age—researchers identified significant correlations with brain health and cognitive decline.
Participants whose speech appeared older than their actual age showed stronger links to Alzheimer’s disease, frontotemporal dementia, and mild cognitive impairment. The findings also aligned with other biological markers, such as epigenetic aging and neuroimaging results. Because this method is non-invasive and low-cost, scientists suggest it could eventually serve as an accessible screening tool for neurodegenerative diseases, particularly in low-resource regions where expensive brain scans or blood tests are less available.
Entities
Adolfo Ibáñez University · Agustín Ibáñez · Global Brain Health Institute · Jed Meltzer · Science Advances · Trinity College Dublin · University of Toronto
Claims
What the coverage asserts, and how many sources carry each claim.
- [● 2 SOURCES] Differences in speech age were most pronounced in participants with Alzheimer's, mild cognitive impairment, and frontotemporal dementia. www.kissfm.es · www.news-medical.net
- [DISPUTED] Researchers developed a ‘speech clock’ that uses machine learning to estimate chronological age from speech characteristics. www.kissfm.es · www.news-medical.net · www.nature.com
- [● 2 SOURCES] A ‘speech age gap’—the difference between predicted and chronological age—is associated with biological aging, brain health, and cognitive decline. www.news-medical.net · www.nature.com
- [○ 1 SOURCE] The tool offers a low-cost, non-invasive alternative to expensive technologies like brain scans or blood tests. www.nature.com
- [○ 1 SOURCE] The tool cannot currently be used to diagnose dementia or predict future cognitive impairment. www.kissfm.es
- [● 2 SOURCES] The study analyzed 2,928 Spanish-speaking participants from Argentina, Chile, Colombia, Mexico, and Peru. www.kissfm.es · www.news-medical.net
- [● 3 SOURCES] Researchers created a ‘speech clock’ using machine learning to estimate chronological age. emvolos.gr · www.ekriti.gr · popaganda.gr
- [● 3 SOURCES] The study analyzed data from 2,928 Spanish-speaking participants. emvolos.gr · www.ekriti.gr · popaganda.gr
- [● 3 SOURCES] A research team led by Trinity College Dublin developed a machine learning model to analyze speech. emvolos.gr · www.ekriti.gr · popaganda.gr
- [● 3 SOURCES] The study sample included healthy adults and individuals with Alzheimer's disease or frontotemporal dementia. emvolos.gr · www.ekriti.gr · popaganda.gr
- [● 3 SOURCES] The discrepancy between chronological age and speech-estimated age is linked to biological aging, brain health, and cognitive function. emvolos.gr · www.ekriti.gr · popaganda.gr
- [● 3 SOURCES] The study was published in the journal Science Advances. emvolos.gr · www.ekriti.gr · popaganda.gr