<?xml version="1.0" standalone="yes"?>
<Paper uid="H01-1016">
  <Title>Development of the HRL Route Navigation Dialogue System</Title>
  <Section position="7" start_page="1" end_page="1" type="concl">
    <SectionTitle>
5. SUMMARY AND FUTURE PLANS
</SectionTitle>
    <Paragraph position="0"> We have reported on our initial implementation and results for an in-vehicle navigation system through the first phase of our system development effort and into the second phase. Full exploitation of the natural language interface is not fully completed in the first phase because we are still developing an operational in-vehicle navigation system to integrate with our dialogue system. The full interface, including destination entry, route planning, position tracking, and map matching will be available later this year. We have, however, developed most of the NL components needed for accessing the database functionality as it comes on-line. We plan to add other important functionality such as points-of-interest and traffic conditions as the project progresses.</Paragraph>
    <Paragraph position="1"> Two other major elements need to be further explored to gain full system functionality. The first is recognizer robustness in the presence of in-vehicle noise during normal everyday use; the second is the street name recognition and pronunciation synthesis problem. Recognizer performance is being addressed by means of a full-scale data collection and corpora development project, in collaboration with GM and the Center for Spoken Language Research at the University of Colorado at Boulder. This work will provide the in-vehicle acoustic data needed to re-train the recognizer models as well as provide a database for developing noise-mitigation and speaker adaptation algorithms for improving recognizer performance. We are developing a method for dynamic loading of street names which we will report on in the near future.</Paragraph>
    <Paragraph position="2"> Acknowledgments. This work was supported in part by a research contract from General Motors.</Paragraph>
  </Section>
class="xml-element"></Paper>