Analysis of conversations in a context of telepresence robot for the elderly

Title:Analysis of conversations in a context of telepresence robot for the elderly
Institute:University of Twente (HMI)
Place:Enschede The Netherlands
Type:Capita selecta and Research Topics
End date:not present
HMI ContactKhiet TruongJaebok Kim


As the number of the elderly is increasing, more intelligent services for them have been investigated and automatic telepresence robots for better social life are becoming more realistic. When we design robots for the elderly, we should consider hearing problems since half of the population aged 75 and above have hearing problems. For socially normative interactions through the robot, the robot should understand whether interaction targets have hearing problems or how severe the hearing problem is. Our final aim is to measure quality of the elderly conversations, mainly affected by hearing problems, and we would like you to analyze conversations collected through the telepresence robot in an automatic way.

We expect the elderly to have different speaking styles depending on levels of hearing problems. As an exploratory study, you investigate characteristics of conversations in terms of non-verbal features, e.g. duration of speech, pitch, energy, overlaps, and turn takings respect to the levels of hearing problems. In this study, we simplify and instantiate hearing problems as “worst case” and “better case” by controlling the speaker volume of the robot or the distance between the robot and the interaction target. For this, we will collect conversations of the elderly and provide you with the corpus. You will analyze the conversations by automatic non-verbal feature extraction tools, which are provided as well. Our research questions are: How do characteristics of nonverbal features vary from different levels of hearing problems? Which nonverbal feature is discriminative between the levels?

Related literature:

Kristoffersson, Annica, et al. "Towards measuring quality of interaction in mobile robotic telepresence using sociometric badges." Paladyn, Journal of Behavioral Robotics 4.1 (2013): 34-48.

Heldner, Mattias, and Jens Edlund. "Pauses, gaps and overlaps in conversations." Journal of Phonetics 38.4 (2010): 555-568.