This post-doc announcement might be of interest to the people on this mailing list who are interested in signal processing. Les is a really good signal-processing theory guy, but he loves auditory processing.
Contact details are below.
A 2 year post-doc is available for US Citizens who have received their Ph.D. within the last 5 years. Also, for highly qualified applicants, current Ph.D. students who will finish within the next year will be considered. If interested, please contact, including your CV, Prof. Les Atlas, Bloedel Research Award Fellow, at atlas@xxxxxx, no later than this June 29, 2012. The Fellowship can possibly start as early as this July 2012 or have its start deferred until Fall 2012 or later, into mid-2013.
This US Government-funded Post-Doctoral Fellowship is based upon Atlas' lab's recent new theory (joint with others, internationally, and with University of Washington Professor Bishnu Atal) and has opportunity for both theoretical and experimental work, especially work which extends the lab's theory and/or raises new science questions. Previously loosely-defined modulation or product models is known to be fundamental to auditory and other aspects of perception. While signal processing tools and science experiments have previously assumed piece-wise stair-step stationary models, better theory, especially for smoothly-changing multivariate dynamic signals at the rates of pitch and syllabic rhythm in speech (and other natural processes), are now available, both statistical and via deterministic generative models. Some applications of these new models include very new approaches to scientific studies and computational models of audio grouping and streaming, better understanding of neural encoding and rhythms, and engineering applications, include talker separation and speech enhancement, such as for hearing assistance devices. Please note that this work does not involve language modeling per se, but could also be used to provide new future tools for language modeling.
This program is highly selective yet allows a chance to meet and collaborate with the best National researchers in areas related to speech, perception, signal processing, brain/cognitive modeling, and biomimetic machine learning. Also, a requirement of this Fellowship will include 1-2 trips a year to Washington, DC to meet with other Post-Doctoral Fellows, and their advisors, who represent the best National research labs along with funding to travel to top research conferences.
While US Citizenship is required, all research under this Fellowship is unrestricted and potentially publishable.
Prof. Les Atlas
Department of Electrical Engineering
University of Washington
Seattle, WA 98195-2500