ECE 537 - Speech Processing Fundamentals

Summer 2009 | Fall 2009 | Spring 2010 | Summer 2010
Section Type Times Days Location Instructor
E LEC 1200 - 1250 M W F   106B3 Engineering Hall  Mark Hasegawa-Johnson

Web Page http://courses.ece.uiuc.edu/ece537/
Official Description Development of an intuitive understanding of speech processing by the auditory system, in three parts. I: The theory of acoustics of speech production, introductory acoustic phonetics, inhomogeneous transmission line theory (and reflectance), room acoustics, the short-time Fourier Transform (and its inverse), and signal processing of speech (LPC/CELP/VQ). II: Psychoacoustics of speech perception, critical bands, masking (JNDs), and the physiology of the auditory pathway (cochlear modeling). III: Information theory entropy, channel capacity, the confusion matrix, state models, EM algorithms, and Bayesian networks. Presentation of classic papers on speech processing and speech perception by student groups. Matlab (or equivalent) programming in majority of assignments. Prerequisite: ECE 410.
Hours 4 hours.
Subject Area Biomedical Imaging, Bioengineering, and Acoustics
Course Prerequisites Credit in ECE 410
Course Directors Jont Allen
Description This course will bring the student to the forefront of signal processing with many practical results and a fundamental understanding of the basic requirements to develop novel algorithms in speech recognition and processing, where the resulting signals are meant for listening, such as speech coding. Speech processing in three parts:
  • I: The theory of acoustics of speech production, introductory acoustic phonetics, including inhomogeneous transmission line theory, reflectance, room acoustics, the short-time Fourier Transform (and its inverse), and signal processing of speech, such as LPC/CELP/VQ.
  • II: Psychoacoustics of speech perception, critical bands, masking JNDs, and the physiology of the auditory pathway; cochlear modeling.
  • III: Information theory, entropy, channel capacity, the confusion matrix, state models, EM algorithms, and Bayesian networks. Classic papers on speech processing and speech perception assigned and presented by student groups.
Credit 4 hours
Course Prerequisites ECE 410 and Matlab (or equivalent) programming is required for many of the homeworks.
Texts J. Flanagan, J. Allen, and M. Hasagawa-Johnson, Speech Analysis Synthesis and Perception, 3rd ed., 2008. (Electronic format provided by instructor.)