Talk
Abstract:
The State of the Art in ASR (and Beyond?)
Steve Young
Engineering Department
University of Cambridge
sjy@eng.cam.ac.uk
This
talk briefly reviews the probabilistic framework of modern HMM
systems and then presents an overview of the refinements needed
to achieve state of the art performance. These include the use
of quinphones, soft-tied decision trees, full variance normalisation,
speaker normalisation, speaker adaptation, model merging, interpolated
language models, and word posterior probability decoding. The
talk concludes by highlighting some specific areas where current
architectures are particularly constraining and suggests possible
future areas of development.
Mathematical
Foundations of Speech Processing and Recognition
2000-2001
Program: Mathematics in Multimedia
|
