With development of automatic speech understanding systems and expansion of the frame of decided problems the using pragmatic information became especially actual. The construction of understanding model is started from a purpose of system and in many respects is defined by a subject domain (SD). One of the most important problems in the field of automatic speech understanding is an account of concepts (terms) of a SD, where terms belong to very restricted dictionary and their meaning has already well coordinated beforehand within the limits of the certain community. The system of automatic speech understanding should generalize correctly concepts (terms), for this purpose it is necessary to use knowledge of concept relations within the framework of the certain SD. In this paper the decision of the problem of accounting concepts hierarchy is proposed by of domain ontology. The ontology is presented as a tree of concepts of a SD in a xml-format that provides ease of modification of the ontology and the dictionary of SD as well. The developed verification method of an ontological subset of a hypothesis of the input utterance provides flexibility of automatic understanding system in concerning the natural language phrases, which contain concepts of a various level of generalization within certain SD, without loss of speed and quality of speech processing.
Cite as: Lee, I.V., Ronzhin, A.L., Karpov, A.A. (2004) Semantic-pragmatic processing of natural language for automatic speech understanding systems. Proc. 9th Conference on Speech and Computer (SPECOM 2004), 488-494
@inproceedings{lee04_specom,
author={I. V. Lee and Andrey L. Ronzhin and Alexey A. Karpov},
title={{Semantic-pragmatic processing of natural language for automatic speech understanding systems}},
year=2004,
booktitle={Proc. 9th Conference on Speech and Computer (SPECOM 2004)},
pages={488--494}
}