5th International Conference on Spoken Language Processing
In a limited domain task (e.g. airline reservation, database retrieval, etc) many robust understanding systems, designed for both speech and text input, have been implemented  based on the Stochastic Conceptual Finite-State paradigm (Semantic Network) or CHRONUS paradigm  (Conceptual Hidden Representation of Natural Unconstrained Speech), which establishes relations between conceptual entities through a probabilistic graph-like structure. The use of this kind of grammar to model semantic information presents limitations, which have been analysed during the implementation of a flexible architecture for a robust information retrieval system, based on the same paradigm . We have tried to solve some of them by integrating a set of conceptual probabilistic and non-probabilistic grammars, which allow certain complexity in the functionality of the application, such as applying non-SQL functions to the results of SQL queries in order to retrieve information not explicitly included in the database, translating certain natural spoken sentences (that would produce difficult embedded queries and therefore more natural queries) without so many restrictions in the relative position of inter-concept relationship.
Bibliographic reference. Colas, Josť / Ferreiros, Javier / Montero, Juan Manuel / Pastor, Julio / Gallardo, Ascension / Pardo, Josť Manuel (1998): "On the limitations of stochastic conceptual finite-state language models for speech understanding", In ICSLP-1998, paper 1095.