<?xml version="1.0" standalone="yes"?>
<Paper uid="W97-1204">
  <Title>Integrating Language Generation with Speech Synthesis Concept to Speech System</Title>
  <Section position="8" start_page="27" end_page="27" type="concl">
    <SectionTitle>
7 Conclusion and Future work
</SectionTitle>
    <Paragraph position="0"> In this paper, a new CTS architecture is presented.</Paragraph>
    <Paragraph position="1"> The key idea is to integrate current NLG and TTS systems in a standard way so that the CTS system developed is able to use any existing NLG tools, STP algorithms and TTS systems and benefit from the information available from NLG. A Speech Integrating Markup Language is designed for this purpose.</Paragraph>
    <Paragraph position="2"> In the future, we will extend our STP algorithms, to predict an adjustable range of speaking rate and stress placement based on discourse, semantic and syntactic information. As a result, we need to extend our SIML so that new information can be incorporated easily.</Paragraph>
  </Section>
class="xml-element"></Paper>