<?xml version="1.0" standalone="yes"?>
<Paper uid="A00-2025">
  <Title>Minimizing Word Error Rate in Textual Summaries of Spoken Language</Title>
  <Section position="11" start_page="190" end_page="190" type="concl">
    <SectionTitle>
9 Summary
</SectionTitle>
    <Paragraph position="0"> In this paper, we presented experiments on summaries of both human and machine generated transcripts from four recordings of spoken language. We explored the trade-off of word accuracy vs. summary accuracy (relevance) using speech recognizer confidence scores to rank passages with lower word error rate higher in the summarization process.</Paragraph>
    <Paragraph position="1"> Results comparing our approach to a simple MMR ranking show that while the WER can be reduced by over 10%, summarization accuracy improves by over 15% as measured against transcripts with relevance annotations.</Paragraph>
  </Section>
class="xml-element"></Paper>