<?xml version="1.0" standalone="yes"?>
<Paper uid="N04-4032">
  <Title>Parsing Conversational Speech Using Enhanced Segmentation</Title>
  <Section position="7" start_page="0" end_page="0" type="concl">
    <SectionTitle>
5 Discussion
</SectionTitle>
    <Paragraph position="0"> In comparison to the na&amp;quot;ive pause-based SU detector, using an SU detector based on prosody and lexical cues gives us more than 45% of the possible gain to the best possible (oracle) case, despite a relatively high SU error rate. We hypothesize that low SU and IP recall in the na&amp;quot;ive segmenter created a much larger parse search space, leading to more opportunities for errors. The improvement is promising, and suggests that research to improve metadata extraction can have a direct impact on the performance of other natural language applications that deal with conversational speech.</Paragraph>
    <Paragraph position="1"> The use of SUs and IPs as input words may result in a loss of information, reducing the &amp;quot;true&amp;quot; word history available to the parser component models. Further research using the structured language model could incorporate these metadata directly into the model, allowing it to take advantage of higher-level metadata without reducing the effective number of words available to the model.</Paragraph>
    <Paragraph position="2"> In addition, just as the SLM is useful for both parsing and language modeling, it could be used to predict metadata for its own sake or to improve word recognition, with or without the word-based representation.</Paragraph>
  </Section>
class="xml-element"></Paper>