In order to do basic research on Multimpdal Dialog between an operator and a computer, we have been developing a multimedia work station named SHIVA. It is built around a PC 386-like micro-computer and drives a lot of different communication media : classical peripherals as keyboard and mouse, and specific devices such as a high definition tactile screen (Elographics), a voice recognition system (DATAVOX-Vecsys). In the fixture, we will add a gesture input device (DataGlove-VPL) and an image video input system (MATROX).
The work presented concerns the study of a multimodal dialog interface based on the use of the tactile screen, the mouse and the voice recognition system. We focus our attention on the "input data" aspect, the feed-back being done through a graphical visualisation. In order to experiment the interface model, we have developed an application concerning the creation and the manipulation of simple graphical objects on the screen.
The aim was to make the multimodality completely (at least, as much as possible) independent from the application. It was therefore necessary to separately define an application model and a user manipulation model.
Cite as: Teil, D., Bellik, Y. (1991) Multimodal dialog interface on a PC-like work station. Proc. 2nd VENACO Workshop - The Structure of Multimodal Dialogue, 175-176
@inproceedings{teil91_smmd,
author={Daniel Teil and Yacine Bellik},
title={{Multimodal dialog interface on a PC-like work station}},
year=1991,
booktitle={Proc. 2nd VENACO Workshop - The Structure of Multimodal Dialogue},
pages={175--176}
}