<?xml version="1.0" standalone="yes"?>
<Paper uid="W90-0122">
  <Title>The Computer Generation of Speech with Dlscoursally and Semantically Motivated Intonation</Title>
  <Section position="2" start_page="0" end_page="164" type="intro">
    <SectionTitle>
2. The Relevant Components of the
</SectionTitle>
    <Paragraph position="0"/>
    <Section position="1" start_page="0" end_page="164" type="sub_section">
      <SectionTitle>
Communal Model
</SectionTitle>
      <Paragraph position="0"> The major components of the overall model that will bC/ referred to below are as follows. We assume an Interactive system, rather than one that is merely monologue. We shall ignore here the components related to parsing, interpretation and inputting to the belief system and planner. The components relevant to generation are:  1. The belief system, which includes general and specific beliefs about ('knowledge of) situations and things in some domain; specific beliefs about the content of the preceding discourse, about vozious aspects of the current social situation, about the addressee and his beliefs of all types, his attitudes, his goals and plans.</Paragraph>
      <Paragraph position="1"> 2. A planner, which makes general plans, drawing on knowledge of ...</Paragraph>
      <Paragraph position="2"> 3. genres (scripts, schemas, etc), introducing  where appropriate sub-units such as transactions (see below) and more detailed plans, using ... 4, the local discourse grammar, which is modelled as a 'systemic flowchart' (i.e. a flowchart containing many small system networks at the choice points, and which generates exchanges and their structure), 5. the lexlcogrammar, i.e. the sentence generator, consisting of:  a. the system networks of semantic features for a wide variety of types of mea-;-g related to situations and realized in the clause, including theme and information structure as well as transitivity, mood, negativity, modallty, affeetive meaning and logical relationships, and equivalent system networks for thiage and qualities, and b. the realization rules which turn the selection expressions of features that are the output from passes through the system networks into syntactic structures with Items (grammatical and lexical) and markers of punctuaUou or intonation as their term;hal nodes.</Paragraph>
    </Section>
  </Section>
class="xml-element"></Paper>