|
Title:
|
PHEASY PHYSICAL EXERCISE ASSISTANCE SYSTEM EVALUATION AND USABILITY STUDY |
|
Author(s):
|
Jan Macek, Jan Kleindienst |
|
ISBN:
|
978-972-8939-18-2 |
|
Editors:
|
Katherine Blashki |
|
Year:
|
2010 |
|
Edition:
|
Single |
|
Keywords:
|
ECA, avatar TTS, dialogue system |
|
Type:
|
Full Paper |
|
First Page:
|
35 |
|
Last Page:
|
42 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
We introduce an interactive system for physical exercise of older people. The system motivates the older person to regular physical activity based on a set of easy exercises in monitored environment. Our system employs multi-modal interfaces including speech synthesis and speech recognition, as well as contactless distance measurement with a sonar. The interactions with the system are realized via body movements and voice dialog. To evaluate the performance of the user we experiment with statistical human activity model to estimate mental and physical state of the user. The system interacts with the user through a talking avatar (ECA-based text-to-speech (TTS)) that coaches the user throughout the set of exercises utilizing the underlying human activity model. At well-defined moments, the user can express his/her preferences on the exercise pace and choose the next steps. The robust estimate of the user's mental and physical state is one of the key challenges. The aim is to support a setup where the older people perform the regular exercises at their home without a direct supervision from care-givers. For initial experiments with the activity model we use three types of classifiers---Naïve Bayes, C4.5, and Bagging---to estimate the users state based on monitored features, users performance history, and his subjective assessment of his current condition. |
|
|
|
|
|
|
