<?xml version="1.0" standalone="yes"?>
<Paper uid="H92-1077">
  <Title>SESSION 12: CONTINUOUS SPEECH RECOGNITION AND EVALUATION II*</Title>
  <Section position="2" start_page="379" end_page="379" type="intro">
    <SectionTitle>
MULTIPLE EVALUATION CONDITIONS
</SectionTitle>
    <Paragraph position="0"> The issue was raised of whether the large number of evaluation conditions was a good idea.</Paragraph>
    <Paragraph position="1"> Doug Paul: The multiple conditions added richness and were appropriate for a pilot and for covering the varied goals of different sites.</Paragraph>
    <Paragraph position="2"> Francis Kubala: The sampling of conditions had worked out well; the number of conditions should probably be narrowed a bit over the next few months.</Paragraph>
    <Paragraph position="3"> Victor Zue: There is a concern over too much splitting of the corpus into multiple parts to support the different</Paragraph>
  </Section>
class="xml-element"></Paper>