<?xml version="1.0" standalone="yes"?>
<Paper uid="W05-0805">
  <Title>Revealing Phonological Similarities between Related Languages from Automatically Generated Parallel Corpora</Title>
  <Section position="7" start_page="39" end_page="39" type="concl">
    <SectionTitle>
8 Conclusions and Future Work
</SectionTitle>
    <Paragraph position="0"> We presented a method to automatically build bilingual pronunciation dictionaries that can be used to reveal phonological similarities between related languages. In general, our similarity scores show that the lexicons of German and Dutch are closer related than German and English. Beside the findings about the relatedness between the two language pairs, we think that the classes might be useful for language learning. An interesting point for future work is to apply the methods developed for the identification of cognates to our bilingual word-lists. Beyond the increase in data, a great challenge is to develop models that can express sound changes on the diachronic level adumbrated in Section 6. We also believe that a slightly modified version of our method can be applied to other related language pairs by using the transcription of morphemes.</Paragraph>
  </Section>
class="xml-element"></Paper>