USING MACHINE LEARNING TO PRODUCE EXPRESSIVE MUSICAL PERFORMANCE

PhD Thesis Proposal Defence


Title: "USING MACHINE LEARNING TO PRODUCE EXPRESSIVE MUSICAL PERFORMANCE"

by

Mr. Siu-Hang Lui


Abstract:

The use of artificial intelligence methods as learning tools has become a 
hot topic in recent years, especially for areas requiring large amounts of 
empirical data such as musicology. Recent research has shown that it is 
possible to represent musical style by appropriate numerical parameters, 
and identify different music styles with inductive machines. It is also 
observed that the music style parameters of a performer are locally and 
globally related to each other. Performers tend to perform music sections 
and motives of similar shapes in similar ways, where music sections and 
motives can be identified by an automatic phrasing algorithm. Based on 
these results, an experiment is proposed for producing expressive music 
from raw quantized music files using machine learning methods like Support 
Vector Machines (SVMs). Experimental result shows that it is possible to 
induce some of a performer’s style by using the music parameters extracted 
from the audio recordings of their real performance.


Date:     		Monday, 6 April 2009

Time:                   12:30p.m.-2:30p.m.

Venue:                  Room 3315
 			lifts 17-18

Committee Members:      Prof. Andrew Horner (Supervisor)
 			Prof. Qiang Yang (Chairperson)
 			Dr. David Rossiter
 			Dr. Chiew-Lan Tai


**** ALL are Welcome ****