Special Topics: Large Vocabulary Recognition
Course No: 506/606, Winter 2012


Izhak Shafran
The Center for Spoken Language Understanding
The Oregon Health & Science University

Course Information
Lectures
References
Schedule
Links

Intimity by Jan Koblasa, image from Statistical Methods for Speech
Recognition, with permission from Frederick Jelinek, MIT Press

Course Information

This is an advanced course for researchers and students working on recognizing a large vocabulary of patterns in sequences of real numbers such as digitized signals. The course will focus on techniques developed recently in automatic speech recognition and will include topics such as: (a) creating compact search graphs using finite state transducers, (b) modeling correlations in multivariate feature space, (b) adapting to domain and speakers, (d) recognizing robustly under noise, and (e) detecting speaker characteristics. The course will consist of a mixture of lectures and paper discussions. Students are also expected to lead paper discussions. The course will be a combination of: Pre-requisite: A graduate level course on automatic speech recognition, statistical pattern recognition, or machine learning. For exceptions, contact the instructor.

Lectures and Readings


References

Note: For recently developed techniques, we will rely on selected papers, which will be provided in required readings.

Schedule

MeetingsThu 1000 hrs - noon
VenueCentral 123
Office hoursBy appointment (request by email)

Links

Relevant Software Tools & Other Resources

This page is maintained by Zak Shafran. Last updated on Feb 4, 2010.