<?xml version="1.0" standalone="yes"?>
<Paper uid="H05-2011">
  <Title>DialogueView: an Annotation Tool for Dialogue</Title>
  <Section position="1" start_page="0" end_page="0" type="abstr">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> There is growing interest in collecting and annotating corpora of language use. Annotated corpora are useful for formulating and verifying theories of language interaction, and for building statistical models to allow a computer to naturally interact with people.</Paragraph>
    <Paragraph position="1"> A lot of annotation tools have been built or are being built. CSLU Toolkit (Sutton et al., 1998) and Emu (Cassidy and Harrington, 2001) are built for words transcription or speech events (such as accent); DAT is built for coding dialogue acts using the DAMSL scheme (Core and Allen, 1997); Nb is built for annotating hierarchical discourse structure (Flammia, 1998); annotation toolkits, such as Mate (McKelvie et al., 2001), AGTK (Bird et al., 2001), and Nite (Carletta et al., 2003), are built for users to create their own tools. In this demo, we will present a novel tool, DialogueView, for annotating speech repairs, utterance boundaries, utterance tags, and hierarchical discourse structure altogether.</Paragraph>
    <Paragraph position="2"> The annotation tool, DialogueView, consists of three views: WordView, UtteranceView, and BlockView.</Paragraph>
    <Paragraph position="3"> These three views present different abstractions of a dialogue, which helps users better understand what is happening in the dialogue. WordView shows the words time-aligned with the audio signal. UtteranceView shows the dialogue as a sequence of utterances. It abstracts away from the exact timing of the words and can even skip words, based on WordView annotations, that do not impact the progression of the dialogue. BlockView shows the dialogue as a hierarchy of discourse blocks, and abstracts away from the exact utterances that were said. Annotations are done at the view that is most appropriate for what is being annotated. The tool allows users to easily navigate among the three views and it automatically updates all views when changes are made in one view.</Paragraph>
    <Paragraph position="4"> DialogueView makes use of multiple views to present different abstractions of a dialogue to users. Abstraction helps users focus on what is important for different annotation tasks. For example, for annotating speech repairs, utterance boundaries, and overlapping and abandoned utterances, WordView provides the exact timing information. For coding speech act tags and hierarchical discourse structure, UtteranceView shows a broader context and hides such low-level details.</Paragraph>
    <Paragraph position="5"> In this presentation, we will show how DialogueView helps users annotate speech repairs, utterance boundaries, utterance tags, and hierarchical discourse blocks. Researchers studying dialogue might want to use this tool for annotating these aspects of their own dialogues. We will also show how the idea of abstraction in DialogueView helps users understand and annotate a dialogue. Although DialogueView focuses on spoken dialogue, we feel that abstraction can be used in annotating monologues, multi-party, and multi-modal interaction, with any type of annotations, such as syntactic structure, semantics and co-reference. Researchers might want to adopt the use of abstraction in their own annotation tools.</Paragraph>
  </Section>
class="xml-element"></Paper>