<?xml version="1.0" standalone="yes"?>
<Paper uid="P03-2034">
  <Title>A speech interface for open-domain question-answering</Title>
  <Section position="7" start_page="0" end_page="0" type="concl">
    <SectionTitle>
6 Conclusion
</SectionTitle>
    <Paragraph position="0"> This paper has described a multimodal interface to a question-answering system designed for rapid input of questions and correction of speech recognition errors. The interface for this demo is Web-based, but should scale to mobile devices. We described a small evaluation of the system's accuracy given raw (uncorrected) transcribed questions from two speakers, which indicates that speech can be used for automatic question-answering, but that an interface for correcting misrecognitions is probably necessary for acceptable accuracy.</Paragraph>
    <Paragraph position="1"> In the future we will continue tightening the integration of the components of the system and port the interface to phones and Palm or Pocket PC devices.</Paragraph>
  </Section>
class="xml-element"></Paper>