Hybrid Statistical and Structural Semantic Modeling for Thai Multi-Stage Spoken Language Understanding
Chai Wutiwiwatchai, Sadaoki Furui · Tokyo Tech Research Repository (Tokyo Institute of Technology) · 2004
This article proposes a hybrid statistical and structural semantic model for multi-stage spoken language understanding (SLU). The first stage of this SLU utilizes a weighted finite-state transducer (WFST)-based parser, which encodes the regular grammar of concepts to be extracted. The proposed method improves the regular grammar model by incorporating a well-known n-gram semantic tagger. This hybrid model thus enhances the syntax of n-gram outputs while providing robustness against speech-recognition errors. With applications to a Thai hotel reservation domain, it is shown to outperform both individual models at every stage of the SLU system. Under the probabilistic WFST framework, the use of N-best hypotheses from the speech recognizer instead of the 1best can further improve performance requiring only a small additional processing time. 1