Language modeling for speech recognizer in dialog systems can take two forms. Human input can be constrained through a directed dialog, allowing the decoder to use a state-specific language model to improve recognition accuracy. Mixedinitiative systems allow for human input that while domainspecific might not be state-specific. Nevertheless, for the most part human input to a mixed-initiative system is predictable, particularly when given information about the immediately preceding system prompt. The work reported in this paper addresses the problem of balancing state-specific and general language modeling in a mixed-initiative dialog system. By incorporating dialog state adaptation of the language model, we have reduced the recognition error rate by 11.5%.
Cite as: Xu, W., Rudnicky, A. (2000) Language modeling for dialog system. Proc. 6th International Conference on Spoken Language Processing (ICSLP 2000), vol. 1, 118-121, doi: 10.21437/ICSLP.2000-29
@inproceedings{xu00_icslp,
author={Wei Xu and Alex Rudnicky},
title={{Language modeling for dialog system}},
year=2000,
booktitle={Proc. 6th International Conference on Spoken Language Processing (ICSLP 2000)},
pages={vol. 1, 118-121},
doi={10.21437/ICSLP.2000-29}
}