In this paper we discuss the role of language modeling in a novel natural language dialogue system designed to automatically route incoming customer calls. We arrive at two significant conclusions: First, standard word error rate measures do not reflect application specific requirements; highly reliable content extraction is possible with relatively high word error rates. Secondly blending human-human data with human-machine data did not improve the performance in language modeling.
Cite as: Reichl, W., Carpenter, B., Chu-Carroll, J., Chou, W. (1998) Language modeling for content extraction in human-computer dialogues. Proc. 5th International Conference on Spoken Language Processing (ICSLP 1998), paper 0588, doi: 10.21437/ICSLP.1998-487
@inproceedings{reichl98_icslp,
author={Wolfgang Reichl and Bob Carpenter and Jennifer Chu-Carroll and Wu Chou},
title={{Language modeling for content extraction in human-computer dialogues}},
year=1998,
booktitle={Proc. 5th International Conference on Spoken Language Processing (ICSLP 1998)},
pages={paper 0588},
doi={10.21437/ICSLP.1998-487}
}