Monitoring Daily Physical Conditions of Older Adults Using Acoustic Features: A Preliminary Result.

Health monitoring in everyday situations has become important due to the rapid aging of many societies. Speech changes have been suggested as a means of measuring an individual's state, such as emotion and stress, and screening for neurodegenerative diseases. However, how speech features are associated with daily physical conditions remains unknown. In this study, we investigated whether acoustic features collected in everyday situations could be used for inferring the daily physical conditions of older adults. We analyzed speech data collected in two settings of monitoring the health of older adults: during phone calls with an actual service for regularly monitoring older adults and with a tablet-based monitoring system we developed. Through analyses, we suggest that acoustic features extracted from speech data in everyday situations may be used for detecting poor physical conditions.