Abstract

We present an approach for achieving scalable authoring of digital learning activities, without sacrificing delivery eloquence. A computer animation character serves as an instructor avatar that not only speaks but also makes deictic, iconic and charisma gestures. The avatar is controlled via a text script, without the prerequisites of computer programming or animation expertise. Given a script for a problem, the system automatically generates scripts for additional instances of the problem, by adapting the targets of the deictic gestures, the speech, and the synchronization between speech and gestures. Starting from initial learning activities of a few minutes, the system can automatically generate hours of quality on-line learning activities. An evaluation by computer graphics, computer animation, and education research experts reveals that the automatically generated learning activities have animation quality that is comparable to that of the original activities.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call