<?xml version="1.0" standalone="yes"?>
<Paper uid="W02-0707">
  <Title>Koji TOCHINAI Graduate school of Business Administration</Title>
  <Section position="6" start_page="0" end_page="0" type="evalu">
    <SectionTitle>
5 Evaluation Experiments
</SectionTitle>
    <Paragraph position="0"/>
    <Section position="1" start_page="0" end_page="0" type="sub_section">
      <SectionTitle>
5.1 Experiments of rule acquisition
</SectionTitle>
      <Paragraph position="0"> All data in experiments are achieved through several speech processes explained in 2.1. Table 2 shows the conditions for experiments. The parameters concerning frame settings have been decided from the results of several preliminary experiments for rule acquisition.</Paragraph>
      <Paragraph position="1">  Many sets of common and different parts were extracted by comparing acoustic characteristics of speech in each language, and translation rules were registered in the translation rule dictionary. Table 3 shows the number of speech utterances and registered translation rules between two languages.</Paragraph>
    </Section>
    <Section position="2" start_page="0" end_page="0" type="sub_section">
      <SectionTitle>
5.2 Experimental results of translation
</SectionTitle>
      <Paragraph position="0"> If an unknown speech utterance of a source language can be replaced with acoustic information from rules in the dictionary, the speech will be translated and synthesized roughly without losing it's meaning.</Paragraph>
      <Paragraph position="1"> Each matched rule includes certain equivalent correspondence parts of the target language. The system needs to decide the most suitable candidates of rules from the rule dictionary for each translation.</Paragraph>
      <Paragraph position="2"> If the level of similarity between the whole applied unknown speech and all parts of the rules is higher than a rate of agreement as in Table 2, the rules that include appropriate parts can become candidates for current translation.</Paragraph>
      <Paragraph position="3"> 82 utterances of limited domain have been applied to the system for translation. Regretfully, we could not obtain any complete translated utterances, although several samples have been incompletely translated by adapting translation rules.</Paragraph>
    </Section>
    <Section position="3" start_page="0" end_page="0" type="sub_section">
      <SectionTitle>
5.3 Discussion
</SectionTitle>
      <Paragraph position="0"> We have to investigate several sources of the experimental results. The rst cause of the failure in the translation can be found in speech data utilized in these experiments. The contents of these utterances do not exactly include the same expression because Table 5: Failures of rule acquisition.</Paragraph>
      <Paragraph position="1"> whole rule the case of the acquisition same content  contents of speech samples are prepared with various ways of speaking even if the semantic information is the same among them.</Paragraph>
      <Paragraph position="2"> Moreover, a small amount of speech data also is another factor because more translation rules should be acquired and adapted for translation.</Paragraph>
      <Paragraph position="3"> The system has performed the task because many suitable rules are registered in the rule dictionary. A sample of parts acquired properly is shown as Table 4. In this table, Japanese words are expressed with an italic font. These parts are successfully acquired through the learning stage, so that many suitable rules can be applied to other unknown speech utterances.</Paragraph>
      <Paragraph position="4"> Therefore, we need to increase the number of speech samples to obtain more translation rules, and it is also necessary to consider the contents of utterances for more effective rule acquisition and application. null In addition, we have paid attention to the parts themselves acquired as translation rules. We have to consider several causes where the same type of sentences is not determined correctly even when the contents are the same. Table 5 shows the number of failures in whole rule acquisition and in the case of comparisons of the same utterances. The types of sentences are determined by the results of the parts extraction stage. In this stage, thresholds have a much important role for deciding common and different parts. Figure 9 shows the distance curves of the same utterances that were not determined as a common part by a threshold. And Figure 10 shows the result of the extraction of common and different parts. Several minimum points of distance curves have been determined as different parts by threshold although two portions of utterances also have the highest similarity in these points. This kind of failure means that the de nition of the threshold has a problem. Therefore, the de nition of the threshold needs to be reconsidered for extracting common and different parts much more correctly.</Paragraph>
    </Section>
  </Section>
class="xml-element"></Paper>