<?xml version="1.0" standalone="yes"?>
<Paper uid="W04-3002">
  <Title>Hybrid Statistical and Structural Semantic Modeling for Thai Multi- Stage Spoken Language Understanding</Title>
  <Section position="1" start_page="0" end_page="0" type="abstr">
    <SectionTitle>
Abstract
</SectionTitle>
    <Paragraph position="0"> 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.</Paragraph>
    <Paragraph position="1"> 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 1-best can further improve performance requiring only a small additional processing time.</Paragraph>
  </Section>
class="xml-element"></Paper>