Keynotes & Tutorials

The development of an automatic speech recognizer is typically a highly supervised process involving the specification of phonetic inventories, lexicons, acoustic and language models, along with annotated training corpora. Although some model parameters may be modified via adaptation, the overall structure of the speech recognizer remains relatively static. While this approach has been effective for problems when there is adequate human expertise and labelled corpora, it is challenged by less-supervised or unsupervised scenarios. It also stands in stark contrast to human processing of speech and language where learning is an intrinsic capability. In this talk I will describe some of the speech and language research topics being investigated in our group that require less conventional expert supervision, and then motivate and describe in more detail some of our efforts to develop speech processing techniques that require fewer or even zero conventional linguistic resources. Time permitting, I hope to describe automatic pronunciation learning via mixture models and their application to organic spoken language interfaces, unsupervised posteriorgram-based spoken term discovery, and an inference-based method to automatically learn phonetic inventories. Short biography James Glass is a Senior Research Scientist at the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL) where he heads the Spoken Language Systems Group. He is also a Lecturer in the Harvard-MIT Division of Health Sciences and Technology. He received his B.Eng. in Electrical Engineering at Carleton University in Ottawa in 1982, and his S.M. and Ph.D. degrees in Electrical Engineering and Computer Science at MIT in 1985, and 1988, respectively. After starting in the Speech Communication group at the MIT Research Laboratory