How to diagnose dementia risk by voice - study
Researchers have developed a system that analyzes over 700 speech features from a short conversation (about 4 minutes) to calculate a "speech age." If it exceeds the actual age, it may indicate cognitive impairment, including dementia. The method was compared with neuroimaging data and Alzheimer's biomarkers.
Researchers have developed a system that evaluates over 700 speech features from a short conversation (about four minutes) and calculates a "speech age." If it exceeds the person's actual age, it may indicate cognitive impairment, including dementia. The algorithm considers word choice, phrase construction, breath control, articulation, tempo, and intonation. Scientists compared data from healthy participants with results from people with mild cognitive impairment, Alzheimer's disease, and various forms of frontotemporal dementia. The speech age gap was found to be associated with signs of brain aging on neuroimaging and epigenetic markers. In Alzheimer's patients, it correlated with elevated levels of p-tau217 in the blood. Analyzing a few minutes of voice could become an accessible method for primary cognitive health screening that can be conducted remotely. The study included 2,928 Spanish-speaking participants from five countries, so the method's accuracy for other languages needs to be verified. For now, the "speech clock" remains an experimental tool.