The technology developed for task-based spoken dialogue systems (SDS) has a significant potential for Computer-Assisted Language Learning. Based on the CMU Let’s Go SDS, we describe two areas in which we investigated adaptations of the technology to non-native speakers: speech recognition and correction prompt generation. Although difficulties remain, particularly towards robust understanding, ...

The paper presents a method for identifying problems of usersystem interaction. Based on a consolidated set of 24 principles of cooperative spoken human-machine dialogue, the paper then proposes and illustrates a general typology of non-cooperative system dialogue behaviour for use in spoken language dialogue analysis and evaluation.

Diagnostic evaluation is an important instrument for the development of high quality spoken language dialogue systems. Yet no rigorous methodology exists for the systematic and exhaustive diagnostic evaluation of all aspects of spoken language interaction: recognition, synthesis, grammar , vocabulary, dialogue etc. The paper addresses part of this problem by presenting a methodology for the dia...

We report on recent work on human-robot spoken dialogue interaction in the context of Hygeiorobot, a project that aims to build a mobile robotic assistant for hospitals. Spoken dialogue systems are particularly suitable to this context, as the robot does not carry a keyboard or other common interaction devices, and is intended to be used by people with little or no computing experience. In this...

We present a Wizard-of-Oz environment for data collection on Referring Expression Generation (REG) in a real situated spoken dialogue task. The collected data will be used to build user simulation models for reinforcement learning of referring expression generation strategies.

We investigate to what extent automatic learning techniques can be used for shallow interpretation of user utterances in spoken dialogue systems. This task involves dialogue act classification, shallow understanding and problem detection simultaneously. For this purpose we train both a rule-induction and a memory-based learning algorithm on a large set of surface features obtained by affordable...

User simulation is an important research area in the field of spoken dialogue systems (SDSs) because collecting and annotating real human–machine interactions is often expensive and time-consuming. However, such data are generally required for designing, training and assessing dialogue systems. User simulations are especially needed when using machine learning methods for optimizing dialogue ma...

We build dialogue system policies for negotiation, and in particular for argumentation. These dialogue policies are designed for negotiation against users of different cultural norms (individualists, collectivists, and altruists). In order to learn these policies we build simulated users (SUs), i.e. models that simulate the behavior of real users, and use Reinforcement Learning (RL). The SUs ar...

Spoken dialogue system performance can vary widely for different users, as well for the same user during different dialogues. This paper presents the design and evaluation of an adaptive version of TOOT, a spoken dialogue system for retrieving online train schedules. Adaptive TOOT predicts whether a user is having speech recognition problems as a particular dialogue progresses, and automaticall...

This paper presents and analyzes an approach to crowd-sourced spoken dialogue data collection. Our approach enables low cost collection of browser-based spoken dialogue interactions between two remote human participants (human-human condition) as well as one remote human participant and an automated dialogue system (human-agent condition). We present a case study in which 200 remote participant...