<?xml version="1.0" standalone="yes"?>
<Paper uid="P91-1038">
  <Title>A Preference-first Language Processor Integrating the Unification Grammar and Markov Language Model for Speech Recognition-ApplicationS</Title>
  <Section position="8" start_page="296" end_page="297" type="concl">
    <SectionTitle>
7. Concluding Remarks
</SectionTitle>
    <Paragraph position="0"> In this paper, we have proposed an efficient language processor for speech recognition applications, in which the unification grammar and the Markov language model are properly integrated in  a preference-first parsing algorithm defined on an augmented chart. Because the unification-based analysis eliminates all illegal combinations and the Markovian probabilities of constituents indicates the correct direction of processing, a very high correct rate of recognition can be obtained. Meanwhile, many unnecessary computations can be effectively eliminated and very high processing speed obtained due to the significant reduction of the huge search space. This preference-first language processor is quite general, in which many different parsing strategies defined by appropriately chosen construction principles and decision rules can be easily implemented for different speech recognition applications.</Paragraph>
  </Section>
class="xml-element"></Paper>