Dealing with Out of Domain Questions in Virtual Characters
We consider the problem of designing virtual characters that support speech-based interactions in a limited domain. Previously we have shown that classification can be an effective and robust tool for selecting appropriate in-domain responses. In this paper, we consider the problem of dealing with out-of-domain user questions. We introduce a taxonomy of out-of-domain response types. We consider three classification architectures for selecting the most appropriate out-of-domain responses. We evaluate these architectures and show that they significantly improve the quality of the response selection making the user’s interaction with the virtual character more natural and engaging.
KeywordsLanguage Model Virtual Character Real Human Intelligent User Interface Human Coder
Unable to display preview. Download preview PDF.
- 2.Leuski, A., Pair, J., Traum, D., McNerney, P.J., Georgiou, P., Patel, R.: How to talk to a hologram. In: Edmonds, E., Riecken, D., Paris, C.L., Sidner, C.L. (eds.) Proceedings of the 11th international conference on Intelligent user interfaces (IUI 2006), Sydney, Australia, pp. 360–362. ACM Press, New York (2006)CrossRefGoogle Scholar
- 3.Leuski, A., Patel, R., Traum, D., Kennedy, B.: Building effective question answering characters. In: Proceedings of the 7th SIGdial Workshop on Discourse and Dialogue (2006)Google Scholar
- 4.Sethy, A., Georgiou, P., Narayanan, S.: Building topic specific language models from webdata using competitive models. In: Proceedings of EUROSPEECH, Lisbon, Portugal (2005)Google Scholar
- 5.Pellom, B.: Sonic: The University of Colorado continuous speech recognizer. Technical Report TR-CSLR-2001-01, University of Colorado, Boulder, CO (2001)Google Scholar