Skip to main content
Aller à la page d’accueil de la Commission européenne (s’ouvre dans une nouvelle fenêtre)
français fr
CORDIS - Résultats de la recherche de l’UE
CORDIS
Contenu archivé le 2024-06-16

Dealing with uncertainty in spoken language processing: Reasoning and problem-management

Objectif

Spoken dialogue systems (SDSs), i.e. computer systems that can understand and react to spoken input and conduct a dialogue about a restricted topic with a human user, have enormous commercial and social potential, by making information and services (customer services, governmental information and services, etc.) accessible via the most natural interface available to people: speech. Present systems, however, are still quite error-prone and often misunderstand their users in various ways.

Our project aims at systematically researching the issue of clarifying misunderstandings, and at enabling systems to intelligently negotiate understanding problems with their users. The host institution, Potsdam University, has built up considerable expertise in symbolic-reasoning approaches to dialogue modelling, but lacks the complementary knowledge of statistical methods and speech processing algorithms and technology. Thus our current research treats these issues as "black boxes'' --- which places significant limitations on the goals that can be achieved. Consequently, our ToK proposal seeks to close this gap by attracting two experienced researchers (for 24 months each) and two more-experienced researchers (for 4 months each) to collaborate with us on the specific project on clarifying misunderstandings, thereby transferring to us knowledge about statistical methods and speech processing, which however is of much broader relevance to us than for the project alone.

We aim at integrating the incoming fellows not only into our research group but also to encourage them to participate in teaching our students. Furthermore, our connections to local companies that market voice dialogue systems are also expected to benefit from the project, as our overall expertise in building dialogue systems grows. In this way, the project can contribute to strengthen the potential of Dialogue Systems research in Berlin/Brandenburg, which is a less-favoured area of Europe.

Champ scientifique (EuroSciVoc)

Ce projet n'a pas encore été classé par EuroSciVoc.
Proposez les domaines scientifiques qui vous semblent les plus pertinents et aidez-nous à améliorer notre service de classification.

Thème(s)

Appel à propositions

FP6-2002-MOBILITY-3
Voir d’autres projets de cet appel

Régime de financement

TOK - Marie Curie actions-Transfer of Knowledge

Coordinateur

UNIVERSITY OF POTSDAM
Contribution de l’UE
Aucune donnée
Coût total
Aucune donnée