<?xml version="1.0" standalone="yes"?>
<Paper uid="W06-1316">
  <Title>Multimodal Dialog Description Language for Rapid System Development</Title>
  <Section position="3" start_page="0" end_page="109" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> In recent years, various types of interactive agents, such as personal robots, life-like agents (Kawamoto et al. 2004), and animated agents are developed for many purposes. Such interactive agents have an ability of speech communication with human by using automatic speech recognizer and speech synthesizer as a main modality of communication. The purpose of these interactive agents is to realize a user-friendly interface for information seeking, remote operation task, entertainment, etc.</Paragraph>
    <Paragraph position="1"> Each agent system is controlled by different description language. For example, Microsoft agent is controlled by JavaScript / VBScript embedded in HTML files, Galatea (Kawamoto et al..</Paragraph>
    <Paragraph position="2"> 2004) is controlled by extended VoiceXML (in Linux version) and XISL (Katsurada et al. 2003) (in Windows version). In addition to this difference, these languages do not have the ability of higher level task definition because the main elements of these languages are the control of modality functions for each agent. These make rapid development of multimodal system difficult. null In order to deal with these problems, we propose a multimodal interaction description language, MIML (Multimodal Interaction Markup Language), which defines dialogue patterns between human and various types of interactive agents by abstracting their functions. The feature of this language is three-layered description of agent-based interactive systems.</Paragraph>
    <Paragraph position="3"> The high-level description is a task definition that can easily construct typical agent-based interactive task control information. The middle-level description is an interaction description that defines agent's behavior and user's input at the granularity of dialogue segment. The low-level description is a platform dependent description that can override the pre-defined function in the interaction description.</Paragraph>
    <Paragraph position="4"> The connection between task-level and interaction-level is realized by generation of interaction description templates from the task level description. The connection between interaction-level and platform-level is realized by a binding mechanism of XML.</Paragraph>
    <Paragraph position="5"> The rest of this paper consists as follows. Section 2 describes the specification of the proposed language. Section 3 explains a process of rapid multimodal dialogue system development. Section 4 gives a comparison with existing multi-modal languages. Section 5 states conclusions and future works.</Paragraph>
  </Section>
class="xml-element"></Paper>