<?xml version="1.0" standalone="yes"?>
<Paper uid="W04-1711">
  <Title>CL for CALL in the Primary School</Title>
  <Section position="7" start_page="0" end_page="0" type="concl">
    <SectionTitle>
6 Conclusion
</SectionTitle>
    <Paragraph position="0"> It is difficult to integrate CL/NLP resources into CALL, especially as these resource generally designed with a CALL audience in However, there are environments wher be successfully integrated, especi imaginative and useful way. The technolog not have to be particularly complex - what is important is that it is appropriately deployed.</Paragraph>
    <Paragraph position="1"> This paper outlined how two NLP resources can be used in the development of CALL resources for primary schools. It is novel in the ICA employ CL/NLP technologies for especially when they are beginners in learning a  young learners, language. We outlined how the output of a FST engine can feed into the generation of Flash animations for Irish verb conjugations. We showed how a POS tagger can be used to annotate a curriculum to produce a corpus which can in turn be used to automatically generate exercises. Both of these initial modules will be comprehensively deployed and evaluated in the classroom during the coming school year (Sept. 2004-June 2005). Future modules will include type A CL technology like DCGs and will take advantage of the controlled languages and limited domains which exist in the primary school environment. Each module of the overall system is being developed so that concurrent evaluation can be carried out.</Paragraph>
    <Paragraph position="2"> This paper highlighted the point that even though neither of these NLP resources was developed with CALL applications in mind, when combined with relatively straightforward programming and interface techniques, they can be used fruitfully in a CALL environment.</Paragraph>
  </Section>
class="xml-element"></Paper>