Using Interactive Tasks to Elicit Natural Dialogue
نویسنده
چکیده
Basic research into the relationship between intonation and speaker’s intentions about syntax and information structure addresses whether, when, and how speakers use prosodic information to signal linguistic and paralinguistic meaning. Speakers use prosody for a range of functions in communication: to mark the difference between immediately relevant vs. background information; to express contrast, contradiction, and correction; and to indicate the intended syntax of ambiguous utterances. In this paper, we review the methodologies used to examine intonation and sentence and discourse structure in language production studies. We focus on design conflicts that arise when experimental techniques need to generate a sufficient number of items for analysis, while simultaneously eliciting utterances that are representative of natural, conversational speech. We discuss the advantages and disadvantages inherent in (1) using spontaneous speech vs. various textbased elicitation techniques, (2) the effect of different levels of detail in instructions to speakers, and (3) the effects of manipulating the experimental environment. In the final sections, we present details of the method used in our own ‘tree decoration project’ as well as some preliminary cross-linguistic speech production data. These data include evidence for regularities in the realization of pitch accent placement and type in English discourse structure, and phonetic evidence for the intonational marking of discourse information status for accented and unaccented lexical items in Japanese.
منابع مشابه
Teaching Robots through Situated Interactive Dialogue and Visual Demonstrations
The ability to quickly adapt to new environments and incorporate new knowledge is of great importance for robots operating in unstructured environments and interacting with non-expert users. This paper reports on our current progress in tackling this problem. We propose the development of a framework for teaching robots to perform tasks using natural language instructions, visual demonstrations...
متن کاملDeep Linguistic Processing for Spoken Dialogue Systems
We describe a framework for deep linguistic processing for natural language understanding in task-oriented spoken dialogue systems. The goal is to create domaingeneral processing techniques that can be shared across all domains and dialogue tasks, combined with domain-specific optimization based on an ontology mapping from the generic LF to the application ontology. This framework has been test...
متن کاملFrom single word to natural dialogue
Spoken language dialogue systems represent the peak of achievement in speech technologies in the 20 th century and appear set to form the basis for the increasingly natural interactive systems to follow in the coming decades. This chapter first presents a model of the task-oriented spoken dialogue system, its multiple aspects and some of the remaining research challenges. In the context of this...
متن کاملA Method for the Early Stages of Interactive System Design using UML and Lean Cuisine+
In interactive system design, models and notations are required for describing user tasks, and for describing the structure of the human-computer dialogue to support these tasks. These descriptions should ideally be linked. This paper examines task modelling in UML and dialogue description in Lean Cuisine+, and describes a method for the early stages of interactive systems design which incorpor...
متن کاملITSPOKE: An Intelligent Tutoring Spoken Dialogue System
ITSPOKE is a spoken dialogue system that uses the Why2-Atlas text-based tutoring system as its “back-end”. A student first types a natural language answer to a qualitative physics problem. ITSPOKE then engages the student in a spoken dialogue to provide feedback and correct misconceptions, and to elicit more complete explanations. We are using ITSPOKE to generate an empirically-based understand...
متن کامل