<?xml version="1.0" standalone="yes"?>
<Paper uid="W04-2302">
  <Title>Stochastic Language Generation in a Dialogue System: Toward a Domain Independent Generator</Title>
  <Section position="9" start_page="0" end_page="0" type="concl">
    <SectionTitle>
7 Conclusion
</SectionTitle>
    <Paragraph position="0"> We have shown that steps toward a domain-independent NLG component of a dialogue system can be taken through a corpus-based approach. By depending on a domain-independent semantic input in combination with a grammar that over-generates possible English utterances and a newswire language model to choose the best, we have shown that it is possible to generate content rich and natural utterances. We report results in a new, richer domain for stochastic generation research and show our approach resulting in only an 11.6% decrease in naturalness when compared to a human speaker.</Paragraph>
  </Section>
class="xml-element"></Paper>