New AI 'Speech Clock' Predicts Biological Age and Dementia Risk from Voice Patterns
A new artificial intelligence (AI) tool has been developed that can estimate a person's chronological age and assess their risk of cognitive problems, including dementia, by analyzing speech patterns. The 'speech clock' identifies over 700 distinct features in voice recordings, such as pauses, speaking speed, pitch, vocabulary, and emotional expression. In a study involving nearly 3,000 Spanish-speaking adults from Latin America, the AI model was trained to predict age based on these vocal characteristics. Researchers found that individuals whose predicted age was significantly higher than their actual age, indicating a larger 'speech-age gap,' were more likely to exhibit cognitive impairment, dementia (including Alzheimer's and frontotemporal dementia), and signs of accelerated biological aging. These individuals also tended to have less favorable social and economic circumstances. The study, published in the journal Science Advances, suggests that speech analysis could offer a non-invasive and cost-effec...
