<?xml version="1.0" standalone="yes"?>
<Paper uid="C88-1082">
  <Title>LINGUISTIC PROCESSING USING A DEPENDENCY STRUCTURE GRAMMAR FOR SPEECH RECOGNITION AND UNDERSTANDING</Title>
  <Section position="6" start_page="406" end_page="406" type="concl">
    <SectionTitle>
5- Conclusion
</SectionTitle>
    <Paragraph position="0"> This paper proposed an efficient linguistic processing strategy for speech recognition and understanding using a dependency structure grammar.</Paragraph>
    <Paragraph position="1"> This grammar suits processing of phrase-order-free languages such as Japanese and processing the result of front-end speech recognition, which is usually erroneous~ This linguistic processing strategy includes bottom-up parsing and a top-down phrase hypotheses predictor. In particular, the bottom-up parser, taking account of the phonetic and linguistic likelihood, greatly increases the accuracy and speed of speech recognition. The predictor reduces the amount of phonetic processing by pre-selecting the phrase hypotheses. The effectiveness of this parser and predictor was shown in speech recognition experiments.</Paragraph>
    <Paragraph position="2"> Future development is include the statistical likelihood of dependency relationships, integration with the statistical phonetic method like Hidden Markov Models, and higher linguistic processing using the semantics and context knowledge.</Paragraph>
    <Paragraph position="3"> Aaknowlmlgment The authors would like to express their appreciation to Dr. Kiyoshi Sugiyama and Dr. Sadaoki Furui, for their invaluable guidance. The authors would also like to thai( to Dr. Kiyohiro Shikano and Shigeki Sagayama for their useful suggestions.</Paragraph>
  </Section>
class="xml-element"></Paper>