This work shows how a dialogue model can be represented as a Partially Observable Markov Decision Process (POMDP) with observations composed of a discrete and continuous component. The continuous component enables the model to directly incorporate a confidence score for automated planning. Using a testbed simulated dialogue management problem, we show how recent optimization techniques are able to find a policy for this continuous POMDP which outperforms a traditional MDP approach. Further, we present a method for automatically improving handcrafted dialogue managers by incorporating POMDP belief state monitoring, including confidence score information. Experiments on the testbed system show significant improvements for several example handcrafted dialogue managers across a range of operating conditions. 1

In recent years, CP-nets have emerged as a useful tool for supporting preference elicitation, reasoning, and representation. CP-nets capture and support reasoning with qualitative conditional preference statements, statements that are relatively natural for users to express. In this paper, we extend the CP-nets formalism to handle another class of very natural qualitative statements one often uses in expressing preferences in daily life – statements of relative importance of attributes. The resulting formalism, TCP-nets, maintains the spirit of CP-nets, in that it remains focused on using only simple and natural preference statements, uses the ceteris paribus semantics, and utilizes a graphical representation of this information to reason about its consistency and to perform, possibly constrained, optimization using it. The extra expressiveness it provides allows us to better model tradeoffs users would like to make, more faithfully representing their preferences. 1.

In developing telephone-based conversational systems, we foresee future systems capable of supporting multiple domains and flexible vocabulary. Users can pursue several topics of interest within a single telephone call, and the system is able to switch transparently among domains within a single dialog. This system is able to detect the presence of any out-of-vocabulary (OOV) words, and automatically hypothesizes each of their pronunciation, spelling and meaning. These can be confirmed with the user and the new words are subsequently incorporated into the recognizer lexicon for future use. This thesis

We present a new approach for adaptive presentation of structured information, based on preference-based constrained optimization techniques rooted in qualitative decisiontheory.

In the past, recurrent networks have been used mainly in neurocognitive or psycholinguistically oriented approaches of language processing. Here we examine recurrent neural networks for their potential in a dicult spoken language classi cation task. This paper describes an approach to learning classi cation of recorded operator assistance telephone utterances. We explore simple recurrent networks using a large, unique telecommunication corpus of spontaneous spoken language. Performance of the network indicates that a semantic SRN network is quite useful for learning classi cation of spontaneous spoken language in a robust manner, which may lead to their use in helpdesk call routing.

As a step toward conversational systems that al- low for a more natural human-computer interac- tion, we repOt'* og GsG, a system that, while pro- viding a natul'al-language interface to a variety of applications, engages in clarification dialogues with the end user through which new semantic mappings are dynamically acquired. GSG exploits task- and language-dependent information but is fully taskand language-independent in its architecture and strategies.

Abstract. Speech interfaces are about to be integrated in consumer appliances and embedded systems and are expected to be used by mobile users in ubiquitous computing environments. This paper discusses some major usability and HCI related problems that may be introduced by this development. It is argued that a human-centered approach should be employed when designing and developing speech interfaces for mobile environments. Further, the Butler, a generic spoken dialogue system developed according to the human-centered approach is described. The Butler features a dynamic multi-domain approach. 1

This paper presents a progress update of our ISIS 1 trilingual spoken dialog system. As described in [8], this is a conversational system for the stocks domain, and supports interactions in the languages of our region – English and two dialects of Chinese (Mandarin and Cantonese). ISIS provides a system test-bed for our initial explorations with the CORBA architecture, and delegation to KQML (Knowledge Query and Manipulation Language) agents. CORBA offers the advantages of interoperability, scalability and location transparency in client/server systems development. Users can delegate tasks to software agents to help monitor information (e.g. a drop in the price of a pre-specified stock), and generate user alert messages. Our current work presents new research directions in the context of ISIS: (i) automatic incorporation of newly listed stocks into our system’s knowledge base; (ii) switching between on-line interaction and off-line delegation in a single dialog thread. We will also report on enhancements in the system’s architecture and features (e.g. automatic end-point detection). delegation in a single dialog thread. We will also report on enhancements in the system’s architecture and features (e.g. automatic end-point detection). 2. System Architecture Previous work in the development of software infrastructures for dialog systems include [11] 2 [2]. Over the past year, we have continued to explore the development of a spoken dialog system based on the CORBA architecture. This middleware resides in between the operating system and the application layer 1.

For telecommunications companies or banks, etc processing spontaneous lanaguage in helpdesk scenarios is important for automatic telephone interactions. However, the problem of understanding spontaneous spoken language is difficult. Learning techniques such as neural networks have the ability to learn in a robust manner. Recurrent networks have been used in neurocognitive or psycholinguistically oriented approaches of language processing. Here they are examined for their potential in a difficult spoken language classification task. This paper describes an approach to learning classification of recorded operator assistance telephone utterances. We explore simple recurrent networks using a large, unique telecommunication corpus of spontaneous spoken language. Performance of the network indicates that a simple recurrent network is quite useful for learning classification of spontaneous spoken language in a robust manner, which may lead to their use in helpdesk call routing. 1

We present a logical approach of spoken language understanding for a human-machine dialogue system. The aim of the analysis is to provide a logical formula, or a conceptual graph, by assembling concepts related to a delimited application domain. This flexible structure is gradually built during an incremental parsing, which is meant to combine syntactic and semantic criteria. Then, a contextual understanding step leads to completing this structure. The evaluations of the current system are encouraging. This approach is a preliminary for a logical dialogue that uses the form of the semantic representations.