Home
Section 800 Tuesday/Thursday 9:00AM - 10:30AM 1014 DOW (Subject to change)
MUSIC SIGNAL PROCESSING
Andrew E. Yagle and Pauline B. Khan
The goals of this course are: (1) To analyze music the way an engineer would analyze an unfamiliar phenomenon; (2) to learn the basics of Fourier signal analysis and synthesis, motivated by observation of musical signals; (3) to apply this to design of two projects; (4) to introduce students to oral and written technical communication, working in teams, and problem solving.
Absolutely no previous knowledge of music is necessary at all. Teams should include a member with some basic knowledge of music and another member with some very basic programming experience. This is NOT a composition or performance arts technology course. No calculus or complex numbers will be required in the course.
Several weekly labs will allow students to compute frequencies of musical notes using signal processing techniques, discover relations between them using semilog plots, learn and then apply basic Fourier signal analysis techniques (including time-frequency analysis) to simple music, and basics of Matlab.
There are two projects: (1) analysis of touch-tone phone signals and Matlab programming of touch-tone analyzers and synthesizers; and (2) Matlab programming of a simple music synthesizer and music transcriber that produces a musical-staff-like transcription in Matlab from .wav files of synthesized music. Both of these projects will requires students to apply their signal processing techniques to analyze, design and test a solution to an open-ended problem.
In the course of these projects, students will learn to function in teams, apply engineering techniques to analysis and design, and write and present technical reports demonstrating and documenting their results. Music transcription is not yet completely solved, so there is a definite open-ended nature to this part of the project. Ethical issues include whether music, regarded as a string of numbers or as a mathematical function, should be copyrightable, and whether a new synthetic instrument should be patentable.
This course will be of interest to students interested in electrical engineering, signal processing, and music technology.
