Asking Questions to Limited Domain Virtual Characters: How Good Does Speech Recognition Have to Be? (bibtex)
by Leuski, Anton, Kennedy, Brandon, Patel, Ronakkumar and Traum, David
Abstract:
In this paper, we describe the evaluation of a limited domain question-answering characters, particularly as to the effect of non-optimal speech recognition, and the ability to appropriately answer novel questions. Results show that answering ability is robust until speech recognition reaches over 60% Word error rate.
Reference:
Asking Questions to Limited Domain Virtual Characters: How Good Does Speech Recognition Have to Be? (Leuski, Anton, Kennedy, Brandon, Patel, Ronakkumar and Traum, David), In Proceedings of the 25th Army Science Conference, 2006.
Bibtex Entry:
@inproceedings{leuski_asking_2006,
	address = {Orlando, FL},
	title = {Asking {Questions} to {Limited} {Domain} {Virtual} {Characters}: {How} {Good} {Does} {Speech} {Recognition} {Have} to {Be}?},
	url = {http://ict.usc.edu/pubs/Asking%20Questions%20to%20Limited%20Domain%20Virtual%20Characters.pdf},
	abstract = {In this paper, we describe the evaluation of a limited domain question-answering characters, particularly as to the effect of non-optimal speech recognition, and the ability to appropriately answer novel questions. Results show that answering ability is robust until speech recognition reaches over 60\% Word error rate.},
	booktitle = {Proceedings of the 25th {Army} {Science} {Conference}},
	author = {Leuski, Anton and Kennedy, Brandon and Patel, Ronakkumar and Traum, David},
	month = nov,
	year = {2006},
	keywords = {Virtual Humans}
}
Powered by bibtexbrowser