Journal
ACM TRANSACTIONS ON ACCESSIBLE COMPUTING
Volume 7, Issue 2, Pages -Publisher
ASSOC COMPUTING MACHINERY
DOI: 10.1145/2744206
Keywords
Alzheimer's disease; mobile robotics; automatic speech recognition; human-computer interaction; smart home
Funding
- Canadian Institutes of Health Research Catalyst Grant: Pilot Projects in Aging
- Natural Sciences and Engineering Research Council of Canada [RGPIN 435874]
- Nuance Foundation
Ask authors/readers for more resources
Increases in the prevalence of dementia and Alzheimer's disease (AD) are a growing challenge in many nations where healthcare infrastructures are ill-prepared for the upcoming demand for personal caregiving. To help individuals with AD live at home for longer, we are developing a mobile robot, called ED, intended to assist with activities of daily living through visual monitoring and verbal prompts in cases of difficulty. In a series of experiments, we study speech-based interactions between ED and each of 10 older adults with AD as the latter complete daily tasks in a simulated home environment. Traditional automatic speech recognition is evaluated in this environment, along with rates of verbal behaviors that indicate confusion or trouble with the conversation. Analysis reveals that speech recognition remains a challenge in this setup, especially during household tasks with individuals with AD. Across the verbal behaviors that indicate confusion, older adults with AD are very likely to simply ignore the robot, which accounts for over 40% of all such behaviors when interacting with the robot. This work provides a baseline assessment of the types of technical and communicative challenges that will need to be overcome for robots to be used effectively in the home for speech-based assistance with daily living.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available