<?xml version="1.0" standalone="yes"?>
<Paper uid="W00-0602">
  <Title>Some Challenges of Developing Fully-Automated Systems for Taking Audio Comprehension Exams</Title>
  <Section position="8" start_page="10" end_page="10" type="concl">
    <SectionTitle>
5 Conclusions
</SectionTitle>
    <Paragraph position="0"> Just as research in reading comprehension can help evaluate text-based NLP systems against a human benchmark, audio comprehension can provide a useful task for evaluating speech understanding systems. Audio comprehension provides an understanding-based evaluation paradigm for speech systems that encourages research on a useful spoken language understanding application rather than on &amp;quot;getting all the words right.&amp;quot; The techniques developed for audio comprehension promise to be widely useful in many language understanding area.</Paragraph>
  </Section>
class="xml-element"></Paper>