Using Machine Learning To Produce Expressive Musical Performance

The Hong Kong University of Science and Technology
Department of Computer Science and Engineering


PhD Thesis Defence


Title: "Using Machine Learning To Produce Expressive Musical Performance"

By

Mr. Siu-Hang Lui


Abstract

The use of artificial intelligence is common in the research of 
musicology, which involves the large-scale analysis of empirical data. 
Recent research studies show that it is possible to represent musical 
style in terms of local and global parame-ters. The local parameters arise 
from the performance of similar motifs in the phrase. The global 
parameters arise from the performance of the phrase as a whole. Performers 
tend to perform similar structures in a similar way. Based on these 
observations, we propose a method for reproducing the style parameters 
from music recordings. The pitch and beat were first extracted using a 
modified algorithm based on Peeter’s and Dixon's algorithms respectively. 
We then tracked the key by Krumhansl-Schmuckler’s algorithm. To predict 
the chord progression,  we used a Hidden Markov Model (HMM) and chord 
transition matrix. To identify the phrases, we segmented the music by 
cadence, recurring pitch patterns, and local energy content. The phrases 
were then trained and re-targeted with a Support Vector Machine (SVM). The 
end result is a reproduction of style parameters including dynamics, tempo 
and articulation. Experiments show that our method reproduces a 
performer's style with a high level of correla-tion to real performances.


Date:			Tuesday, 14 September 2010

Time:			2:00pm – 4:00pm

Venue:			Room 3401
 			Lifts 17/18

Chairman:		Prof. Ping Gao (CBME)

Committee Members:	Prof. Andrew Horner (Supervisor)
 			Prof. David Rossiter
 			Prof. Qiang Yang
                       	Prof. Matthew McKay (ECE)
                         Prof. Kin-Hong Wong (Comp. Sci. & Engg., CUHK)


**** ALL are Welcome ****