In this paper, we introduce an agent-based error handling architecture for spoken dialogue systems. In this architecture, all the parts of the error-handling process on the different interaction levels (input, dialogue and output) are explicitly modeled. Error handling is divided into individual, preferably application independent components. The proposed architecture makes it possible to construct adaptive and reusable error handling components and entire error-handling toolkits. The architecture is especially suitable for multilingual applications. The architecture is implemented as part of the Jaspis speech application development environment and it uses Jaspis' agent-based interaction model.
Cite as: Turunen, M., Hakulinen, J. (2001) Agent-based error handling in spoken dialogue systems. Proc. 7th European Conference on Speech Communication and Technology (Eurospeech 2001), 2189-2192, doi: 10.21437/Eurospeech.2001-516
@inproceedings{turunen01_eurospeech,
author={Markku Turunen and Jaakko Hakulinen},
title={{Agent-based error handling in spoken dialogue systems}},
year=2001,
booktitle={Proc. 7th European Conference on Speech Communication and Technology (Eurospeech 2001)},
pages={2189--2192},
doi={10.21437/Eurospeech.2001-516}
}