We introduce in this paper a hierarchical hybrid statistical language model, represented as a collection of local models plus a general model that binds together the local ones. The model provides a unified framework for modelling language both above and below the word level, and we exemplify with models of both kinds for a large vocabulary task domain. To our knowledge this is the first paper to report an extensive evaluation of the improvements achieved from the use of local models within a hierarchical framework in comparison with a conventional word-based trigram model.
Cite as: Galescu, L., Allen, J. (2000) Evaluating hierarchical hybrid statistical language models. Proc. 6th International Conference on Spoken Language Processing (ICSLP 2000), vol. 2, 963-966, doi: 10.21437/ICSLP.2000-431
@inproceedings{galescu00b_icslp,
author={Lucian Galescu and James Allen},
title={{Evaluating hierarchical hybrid statistical language models}},
year=2000,
booktitle={Proc. 6th International Conference on Spoken Language Processing (ICSLP 2000)},
pages={vol. 2, 963-966},
doi={10.21437/ICSLP.2000-431}
}