PhD Qualifying Examination


"Using Machine Learning to Produce Expressive Musical Performance"

Mr. Siu Hang Lui


Abstract:

The use of artificial intelligence methods as learning tool in research
has become a hot topic in recent years, especially for those have large
amounts of empirical data of highly complex structure, for example,
research in musicology. Recent research has shown that it is possible to
represent music style by proper digital parameters, and it is possible to
identify different music styles by inductive machine, where the machine is
trained with labeled music pieces. Based on these research results, I
propose an experiment to produce expressive music by imposing trained
expressive music parameters into quantized digital music files, using
common machine learning methods like Support Vector Machines (SVMs).


Date:     		Wednesday, 31 January 2007

Time:                   2:00p.m.-4:00p.m.

Venue:                  Room 3494
			lifts 25-26

Committee Members:      Prof. Andrew Horner (Supervisor)
			Dr. Qiang Yang (Chairperson)
			Dr. Chiew Lan Tai
			Dr. Amine Bermak (ECE)


**** ALL are Welcome ****