Search

A common convention in graphical user interfaces is to indicate a “wait state”, for example while a program is preparing a response, through a changed cursor state or a progress bar. What should the analogue be in a spoken conversational system? To address this question, we set up an experiment in which a human information provider (IP) was given their information only in a delayed and incremental manner, which systematically created situations where the IP had the turn but could not provide task-related information. Our data analysis shows that 1) IPs bridge the gap until they can provide information by re-purposing a whole variety of task- and grounding-related communicative actions (e.g. echoing the user’s request, signaling understanding, asserting partially relevant information), rather than being silent or explicitly asking for time (e.g. “please wait”), and that 2) IPs combined these actions productively to ensure an ongoing conversation. These results, we argue, indicate that natural conversational interfaces should also be able to manage their time flexibly using a variety of conversational resources.

Recent advances in incremental language processing for dialogue systems promise to enable more natural conversation between humans and computers. By analyzing the user's utterance while it is still in progress, systems can provide more human-like overlapping and backchannel responses to convey their level of understanding and respond more quickly. In this paper, we look at examples of several overlapping response types in human-human dialogues, and present an initial computational model of the incremental grounding process in these responses. Additionally, we describe an implementation of this model in a virtual human dialogue system that can provide backchannels, head nods, frowns, completions and low latency responses.

The contribution of this dissertation is to show how interlocutors in dialogue can reason probabilistically about natural language interpretation, dialogue state (context), and natural language generation in a way that is consistent with three fundamental claims made by mainstream theories of pragmatic reasoning in human-human dialogue: 1. interlocutors track and exploit the evolving context to coordinate their individual contributions; 2. the current context depends on what the previous utterances of both interlocutors have meant (contributed); 3. what a speaker can recognizably mean (contribute) by a specific choice of words depends on the current context.Mainstream pragmatic theories depend on these assumptions to explain how a speaker can make linguistic choices that the hearer will interpret as intended, but these theories do not lend themselves to straightforward probabilistic reasoning. Engineering approaches to building dialogue systems implement straightforward probabilistic reasoning, but sacrifice one or more (sometimes all) of these fundamental aspects of pragmatic theory in order to do so. This dissertation shows how we can achieve the robustness and data-driven methodology enjoyed by engineering approaches while keeping our interlocutors on a sound theoretical footing, and thereby points the way toward a new class of dialogue systems that are empirically driven, that are robust pragmatic reasoners, and that exhibit human-like sensitivity to the ins and outs of language use in context.

Received views of utterance context in pragmatic theory characterize the occurrent subjective states of interlocutors using notions like common knowledge or mutual belief. We argue that these views are not compatible with the uncertainty and robustness of context-dependence in human–human dialogue. We present an alternative characterization of utterance context as objective and normative. This view reconciles the need for uncertainty with received intuitions about coordination and meaning in context, and can directly inform computational approaches to dialogue.