> College of Engineering > Electrical & Computer Engineering
EE578 Speech Signal Processing
Course Home Page (2008)
Course Description (as described in the MSU Bulletin):
This course is an introduction to speech signal processing. Topics include production & classification of speech sounds, pole-zero models, homomorphic signal processing, short-time Fourier transform analysis and synthesis, filter-bank analysis and synthesis, sinusoidal analysis and synthesis, pitch estimation, and speech coding. Topics related to speech recognition and synthesis
will also be covered.
Course Description and Objective
(as it will most likely unfold in this course instantiation):
The objective of the course is to develop an understanding of how speech signals are processed in three general areas: Analysis, Synthesis, and Recognition. Speech must also be understood
in the context of its creation (anatomy, classification of sounds, etc.) as well as in its perception (psychology & neuroscience). Analytical tools are needed for analysis and synthesis, which
draw on the areas of digital signal processing and time-frequency analysis. Pattern recognition concepts are needed for speech recognition. Finally, since computers cannot process and understand speech as well as humans do, we will look to biology for inspiration since the brain does an amazing job in all these tasks.
The course, in addition to the usual homework and exams, will require a paper to be written (and presented in class) as well as a project that will require programming in Matlab (that will also be
presented in class).