<?xml version="1.0" standalone="yes"?>
<Paper uid="P98-1107">
  <Title>A Method for Correcting Errors in Speech Recognition Using the Statistical Features of Character Co-occurrence</Title>
  <Section position="1" start_page="0" end_page="0" type="abstr">
    <SectionTitle>
Abstract
</SectionTitle>
    <Paragraph position="0"> It is important to correct the errors in the results of speech recognition to increase the performance of a speech translation system. This paper proposes a method for correcting errors using the statistical features of character co-occurrence, and evaluates the method.</Paragraph>
    <Paragraph position="1"> The proposed method comprises two successive correcting processes. The first process uses pairs of strings: the first string is an erroneous substring of the utterance predicted by speech recognition, the second string is the corresponding section of the actual utterance. Errors are detected and corrected according to the database learned from erroneous-correct utterance pairs. The remaining errors are passed to the posterior process which uses a string in the corpus that is similar to the string including recognition errors.</Paragraph>
    <Paragraph position="2"> The results of our evaluation show that the use of our proposed method as a post-processor for speech recognition is likely to make a significant contribution to the performance of speech translation systems.</Paragraph>
    <Paragraph position="3"> method also obtains reliably recognized partial segments of an utterance by cooperatively using both grammatical and n-gram based statistical language constraints, and uses a robust parsing technique to apply the grammatical constraints described by context-free grammar (Tsukada et aL, 97). However, these methods do not carry out any error correction on a recognition result, but only specify correct parts in it.</Paragraph>
    <Paragraph position="4"> In this paper we therefore propose a method for correcting errors, which is characterized by learning the trend of errors and expressions, and by processing in an arbitrary length string.</Paragraph>
    <Paragraph position="5"> Similar work on English was presented by (E.K.</Paragraph>
    <Paragraph position="6"> Ringger et al., 96). Using a noisy-channel model, they implemented a post-processor to correct word-level errors committed by a speech recognizer.</Paragraph>
  </Section>
class="xml-element"></Paper>