We describe an implemented spoken-language dialogue system for a travel-planning domain, which accesses a commercially available travelinformation web-server and supports a flexible mixed-initiative dialogue strategy. We argue, based on data from initial Wizard-of-Oz experiments, that mixed-initiative strategies are appropriate for many types of user, but require more sophisticated architectures for processing of language and dialogue; we then use these observations to motivate an architecture which combines parallel deep and shallow natural language analysis engines and an agenda-driven dialogue manager. We outline the top-level processing strategy used by the dialogue manager, and also a novel formalism, which we call Flat Utterance Description, that allows us to reduce the output of the deep and shallow languageprocessing engines to a common representation.

This paper discusses some of the implicit assumptions made by shallow and deep processing approaches, and advocates a new approach which keeps the robustness of shallow or keyword-based approaches, whilst retaining the generality and formal nature of a full semantics.

We present RUBISC, a new incremental chunker that can perform incremental slot filling and revising as it receives a stream of words. Slot values can influence each other via a unification mechanism. Chunks correspond to sense units, and end-of-sentence detection is done incrementally based on a notion of semantic/pragmatic completeness. One of RU-BISC’s main fields of application is in dialogue systems where it can contribute to responsiveness and hence naturalness, because it can provide a partial or complete semantics of an utterance while the speaker is still speaking. The chunker is evaluated on a German transcribed speech corpus and achieves a concept error rate of 43.3 % and an F-Score of 81.5. 1

The development of spoken dialogue systems is often limited by the performance of their speech recognition component. The impact of speech recognition errors on dialogue systems is often studied at the global level of task completion. In this paper, we carry an empirical study on the consequences of speech recognition errors on a fully-implemented dialogue prototype, based on a speech acts formalisms. We report the impact of speech recognition errors on speech act identification and discuss how standard control mechanisms can participate to robustness by assisting the user in repairing the consequences of speech recognition errors.

In a typical task oriented dialogue system, the interpretation task consists of mapping from the acoustic input to a series of moves. In the simplest cases each move is just a pairing of slots and values e.g. "destination = paris". In this paper we will describe a system for language interpretation which is designed to work with lattice based output from a class based statistical language model. The work described enables phrase and keyword spotting to be incorporated within a uniform approach which also allows information from higher-level linguistic structure to be used if it is both available and likely to be helpful. The work described here is part of a larger collaborative effort being pursued as part of the EU projects D'Homme [1] and Siridus [2] where one thread of the research is hoping to provide better grounds for some of the choices that need to be made for recogniser language modelling and language processing for a variety of spoken dialogue scenarios

As a part of the VR-tual Pharmacy [23], a vision of the future virtual customer service of the Swedish Pharmacy Corporation, Apoteket AB, there will be an automated personal assistant. It shall help the customers to find information and be able to answer questions not needing expert advice. This assistant shall preferably communicate using Swedish natural language. This thesis aims to give an introduction to Natural Language Processing (NLP) and an overview of available technology suitable for implementing the NLP functionality of this assistant. The thesis shall serve as an information resource in the VR-tual Pharmacy project. First, the appropriate theories are presented, then a few existing commercial NLP systems are evaluated and finally a demonstration system is designed and implemented, using agent oriented design and open source NLP libraries. This system shows some of the strengths of NLP dialog systems in customer-service interaction. The Discussion section addresses some shortcomings of the demonstration system such as handling of spelling errors.

A syntax tree or standard semantic representation can be represented as a set of indexed constraints. This paper describes how this idea can be used in task oriented dialogue systems to provide interpretation rules which incorporate structural and contextual constraints where available, and degrade gracefully on ungrammatical input. 1 Introduction This paper applies the idea of distributed representation to the interpretation of dialogue utterances. The approach provides the robustness you might expect from keyword/phrase spotting, with the ability to use higher level structural information where this can provide a more accurate analysis. It is particularly aimed at flexible dialogue systems where user utterances may be long and not necessarily grammatical. In this situation the system needs to extract out as much relevant information as possible even if it cannot provide a grammatical analysis for the utterance as a whole. As an example task we chose the route planning domai...