Welcome to the International Conference on Auditory-Visual Speech Processing (AVSP) 2009. This year AVSP will be held at the University of East Anglia in Norwich. The University is set in 30 acres of parkland on the Western outskirts of the city of Norwich, approximately 2 miles from the city centre. Norwich itself is the regional capital of East Anglia and the city is steeped in history. Many of the medieval buildings have survived the passage of time and are still in use today. Indeed in and around the city centre is an eclectic mix of the old and new, with modern structures, such as The Forum, only a stones throw from the beautiful Elm Hill, a cobbled lane from the Tudor period. Also within the city is Norwich Castle, which dates back to the Norman period, the only permanent outdoor covered market in the UK, and Tombland, home to one of two cathedrals in Norwich.
These proceedings contain the 36 papers accepted for presentation at AVSP 2009, and include four extended abstracts for systems entered into the second LIPS visual speech synthesis challenge. We thank Microsoft Research for providing the use of their Conference Management Service, which was used to manage the paper submission and reviewing process. The reviewing process required that each paper receive at least two reviews, and we are grateful to all of the reviewers listed overleaf for providing their valuable time.
We would like to thank our colleagues at the University of East Anglia who have helped organise the conference, especially Sue Koria for overseeing the processing of registration forms. We are also grateful to the organisers of previous AVSP meetings, in particular Roland Goecke (AVSP 2008) and Jean Vroomen (AVSP 2007), and to Denis Burnham for their valuable advice.
In our programme we have two excellent invited speakers: Professor Peter Bull from the University of York and Professor Louis Goldstein from the University of Southern California. For our keynote speakers, we wanted people who not necessarily are in the mainstream of audiovisual speech research, but have sufficient overlap to stimulate the audience and perhaps spark discussion and broaden the scope of AVSP. We are delighted that our two speakers have accepted our invitation.
We are pleased that AVSP 2009, as with previous years, has attracted a truly interdisciplinary group of delegates. We have papers relating to speech and gesture, speech perception, audiovisual speech synthesis, audiovisual speech recognition and automated lip-reading. This, we are sure, will provide a stimulating set of talks for all.
We hope that you have an enjoyable conference and find the time to visit some of the wonderful sites in and around Norwich.