The Persona system integrates speech recognition, natural language understanding, animation and speech synthesis to create a conversational assistant that interacts with the user in a spoken dialogue. A key goal of the system is to allow users maximum flexibility to express their requests in whatever syntax they find most natural. Therefore, we have chosen to base the interface on a broad-coverage natural language processing system, even though the assistant currently understands requests in only a very limited domain. This paper describes the integration of natural language processing into Persona, indicates why we believe our approach can be extended to new application domains, and argues that the use of a general purpose natural language understanding system is a better strategy for conversational interfaces than the creation of specialized application languages.
Cite as: Ball, J.E., Ling, D.T. (1995) Spoken language processing in the persona conversational assistant. Proc. ESCA Workshop on Spoken Dialogue Systems, 109-112
@inproceedings{ball95_sds,
author={J. Eugene Ball and Daniel T. Ling},
title={{Spoken language processing in the persona conversational assistant}},
year=1995,
booktitle={Proc. ESCA Workshop on Spoken Dialogue Systems},
pages={109--112}
}