<?xml version="1.0" standalone="yes"?>
<Paper uid="W04-1613">
  <Title>Letter-to-Sound Conversion for Urdu Text-to-Speech System</Title>
  <Section position="5" start_page="3" end_page="3" type="concl">
    <SectionTitle>
5 Conclusion
</SectionTitle>
    <Paragraph position="0"> This paper briefly discusses the architecture of Natural Language Processing portion of an Urdu Text-to-Speech system. It explains the details of Urdu consonantal and vocalic system and Urdu letters. Urdu shows regular behavior and thus the phonemic forms are predictable from the textual input. The letter-to-sound rules define this  and not nullnullnull null ) indicating it behaves both like a character and an aerab. It is still unclear on how this behavior is distributed and whether it is predictable. As it is a script centric issue, it is not discussed further here.</Paragraph>
    <Paragraph position="1"> mapping and are thus essential for developing Urdu TTS.</Paragraph>
  </Section>
class="xml-element"></Paper>