Professor Steve Young
Professor Steve Young is pleased to consider applications from prospective PhD students.
http://mi.eng.cam.ac.uk/~sjy/in... (personal home page)
My primary interest is in adaptive learning in spoken dialogue systems. This currently involves modelling human machine dialogues using the framework of Partially Observable Markov Decision Processes (POMDPs). The research issues addressed include user simulation to drive reinforcement learning, scaling POMDPs to make them tractable in real-world problems, tight integration of ASR and semantic decoding to provide noise robustness and rapid on-line adaptation. In the future, it is hoped to further investigate the behaviours learnt by these systems and how they relate to and differ from observed human behaviours.
Associated News Items
Williams JD, Young SJ (2007), “Partially Observable Markov Decision Processes for Spoken Dialog Systems” Computer Speech and Language 21(2):231-422
Young SJ, Schatzmann J, Weilhammer K, Ye H (2007), “The Hidden Information State Approach to Dialog Management” ICASSP 2007:Honolulu, Hawaii
Young SJ (2006), “Using POMDPs for Dialog Management. IEEE/ACL Workshop on Spoken Language Technology (SLT 2006)” Aruba