University of Sheffield

The NLP/SpandH conjunction is the largest in the UK, and specialises in dialogue interfaces as well as a range of topics, such as information extraction, information management, ontology induction, and speech recognition areas, in which it has achieved some of the best scores internationally in US DARPA competitions. Its expressive dialogue work dates back to 1997, when members of the NLP group designed the CONVERSE system that entered the Loebner Competition in New York for the most plausible computer dialogue partner of the year and won. They then began to organise (with EC funding) a series of international workshops on dialogue at Bellagio that brought together industrial and academic work on machine dialogue. The group also participates in a range of EPSRC projects (the most prestigious form of research funding in the UK) including AKT, a five-University six-year interdisciplinary research consortium. SpandH are scientific ccrdinators of AMI, a 6FP Integrated project, on understanding multispeaker meetings.

The Computer Graphics Research Group (see Computer Graphics)is long established, having amongst its members Alan Watt, the author of many famous 3D graphics books. With the acquisition of the new immervise Virtual Reality lab and the facilities at the North Campus has taken a new direction focusing on the creation of Virtual Environments, including the creation of virtual animated characters thier integration with computer dialogue systems.

Yorick Wilks is head of the NLP group and the author of 5 books and hundreds of papers in the area. His Cambridge degrees were first in mathematics and philosophy, and then a PhD in computing and philosophy. He has been a researcher at Cambridge, Stanford, Edinburgh, Essex, New Mexico, Oxford and Sheffield Universities. He is a Fellow of both the European and American Artificial Intelligence societies, a Fellow of the EPSRC College of Computing and a member of the UK Computing Research Council. He designed the CONVERSE system, the VIEWGEN belief and intention modelling system, and the COMPANIONS paradigm of an intelligent and affective agent, designed as a permanent but computer-based personality functioning as a companion to a particular person.

Daniela Romano is a lecturer of the Computer Graphics research group and has obtained a Ph.D. from the University of Leeds, UK, and an MSc in Computer Science University from the University of Bari, Italy. Her career in computer science started with six years in the software development for private sectors companies including Microsoft S.p.A. and the Ford Motor Company) in Italy, Germany and the UK. Daniela’s academic career started in 1996. Before joining the Multimedia Graphics Group at Sheffield, she worked for the Universities of Leeds, University College London and Salford. Daniela specialised in the creation of educational virtual environments for training or therapeutic purposes and in particular in the creation of empathic animated characters, diplaying emotions and personality.

Marilyn Walker is Royal Society Wolfson Professor in Computer Science at the University of Sheffield. She has an M.S. in Computer Science from Stanford University, and an M.A. in Linguistics and a Ph.D. in Computer Science from the University of Pennsylvania. Her research focuses on spoken dialogue systems, specifically on spoken language generation and the use of machine learning in dialogue. As a member of the Speech and Information Processing Lab at AT&T Bell Labs, she built the first dialogue system to use reinforcement learning to adapt the dialogue manager on the basis of interaction with human users, and developed the first trainable generation component for a spoken dialogue system as part of her work on the DARPA sponsored AT&T Communicator project. Her research on spoken language generation has examined the effect of social relationship and cognitive factors on language production.

Roger Moore took up a Chair in SpandH in August 2004 having been Chief Scientific Officer of 20/20 Speech Ltd. since 1999. Prior to that, he was Head of the UK Government's 'Speech Research Unit' (SRU) from 1985. He has a BA in Computer & Communications Engineering, and an MSc and PhD in Automatic Speech Recognition from Essex University. He has authored over 100 scientific publications in Speech Technology algorithms, applications and assessment. He is Editor of Computer Speech & Language, a member of the Editorial Board for Speech Communication, a Fellow of the UK Institute of Acoustics and a Visiting Professor in the Department of Phonetics and Linguistics at University College London. He was President of the 'European/International Speech Communication Association' (ISCA) from 1997 to 2001.