<?xml version="1.0" standalone="yes"?>
<Paper uid="W04-3012">
  <Title>Word level confidence measurement using semantic features. In Proceedings of ICASSP, Hong Kong, April.</Title>
  <Section position="7" start_page="0" end_page="0" type="concl">
    <SectionTitle>
6 Conclusions
</SectionTitle>
    <Paragraph position="0"> In this paper, we presented a system which determines domains of speech recognition hypotheses. Our approach incorporates high-level semantic knowledge encoded in a domain model of ontological concepts. We believe that this type of semantic information has the potential to improve the performance of the automatic speech recognizer, as well as other components of spoken language processing systems.</Paragraph>
    <Paragraph position="1"> Basically, information about the current domain of discourse is a type of contextual knowledge. One of the future challenges will be to find ways of including this high-level semantic knowledge into SLP systems in the most beneficial way. It remains to be studied how to integrate semantic processing into the architecture, including speech recognition and discourse processing.</Paragraph>
    <Paragraph position="2"> An important aspect of the scalability of our methods is their dependence on concept-based domain models. A natural extension would be to replace hand-crafted ontological concepts with, e.g., WordNet concepts. The structure of WordNet can then be used to determine high-level domain concepts that can replace human domain annotations.</Paragraph>
    <Paragraph position="3"> One of the evident problems with this approach is, however, the high level of lexical ambiguity of the WordNet concepts. Apparently, the problem of ambiguity scales up together with the coverage of the respective knowledge source.</Paragraph>
    <Paragraph position="4"> Another remaining challenge is to define the methodology for the evaluation of methods such as proposed herein. We have to think about appropriate evaluation metrics as well as reference corpora. Following the practices in other NLP fields, such as semantic text analysis (SENSEVAL), message and document understanding conferences (MUC/DUC), it is desirable to conduct rigourous large-scale evaluations. This should facilitate the progress in studying the effects of individual methods and cross-system comparisons.</Paragraph>
  </Section>
class="xml-element"></Paper>