This paper presents a description of the dialogue model structured into a hierarchy of sub-dialogues, each one having a particular functional value. The dialogue model has been formalized by the use of formal grammars and speech acts theory : the grammar non-terminal terms correspond to sub-dialogues, and terminals correspond to dialogue acts. We have implemented the dialogue model by transforming the chosen grammar into a finite state network, in which exchanges between the user and the machine correspond to transitions in the network. We describe to in this paper the task model and the functional architecture of the dialogue manager which controls all the dialogue components. The dialogue model has been evaluated in the French version of ATIS (Air Travel Information Services) task by integrating the dialogue system with a speaker independent continuous speech recognizer.
Cite as: Bennacef, S.K., Néel, F., Maynard, H.B. (1995) An oral dialogue model based on speech acts categorization. Proc. ESCA Workshop on Spoken Dialogue Systems, 237-240
@inproceedings{bennacef95_sds,
author={S. K. Bennacef and Françoise Néel and H. B. Maynard},
title={{An oral dialogue model based on speech acts categorization}},
year=1995,
booktitle={Proc. ESCA Workshop on Spoken Dialogue Systems},
pages={237--240}
}