Course Description

Music information retrieval (MIR) is an emerging field of research that aims to provide intelligent solutions for music listening, performance, composition and other musical activities. The basic goal of MIR is inferring various types of information from music, for example, symbolic information such as notes, rhythm and harmony, or semantic information such as genre, mood and other text descriptions of music. This information can be applied to music search or recommendation, music education, entertainment, interactive performance, visualization and so on. This course will introduce various topics in MIR research focusing on computational methods to extract such information from audio files.

General Information

Instructor: Juhan Nam (남주한)
TA: Dasaem Jung (정다샘)
Lecture: Tuesday, Thursday 2:30-4:00
Room: N25#3239 Laughlin Hall (러플린홀)

Grading Policy

3 Assignments (via KLMS) : 60 %
Final Project (presentation and report): 40%

Textbook

Fundamentals of Music Processing, Meinard Müller
http://www.springer.com/gp/book/9783319219448

Schedule

( This schedule is subject to change. )
DateTopics Textbook
Mar 3 Introduction to Music Information Retrieval [slides]
Mar 8 Human Auditory System and MIR System [slides]
Chapter 1
Mar 10 Discrete Fourier Transform (DFT) [slides]
Chapter 2
Mar 15 / 17 Time-Frequency Representations of Audio [slides]
Chapter 2
Mar 22 Assignment #1 due date: Apr 5
Mar 22 / 24 Monophonic Pitch Detection [slides]
YIN paper
Mar 29 / 31 Timbre and Low-Level Audio Features [slides]
Apr 7 / 12 Rhythm Analysis: Onset Detection and Tempo Estimation [slides]
Chapter 6
Apr 12 Assignment #2 Due date: Apr 26
Apr 14 / 19 Tonal Analysis: Chroma Representation, Key and Chord Detection [slides] Chapter 5
Apr 21 / 26 Mid-term Break
Apr 28 Structure Analysis: Music Form and Self-Similarity Matrix [slides] Chapter 4
May 3 Music and Audio Alignment and Dynamic Programming [slides] Chapter 3
May 5 Holiday - No class
May 10 Assignment #3 Due date: May 24
May 10 / 12 Machine Learning and Music Classification [slides] Chapter 5
May 17 / 19 Non-negative Matrix Factoriztion (NMF) and Musical Applications (TA: Dasaem Jeong) [slides] Chapter 8
May 24 Deep Learning for Music Classification (Invited Speaker: Keunwoo Choi from Queen Mary University of London) [slides]
May 26 / 31 Machine Learning and Music Classification (Cont'd)
Jun 2 Music Recommendation [slides] Chapter 7
Jun 7 / Jun 9 TBD
Jun 14 Final Presentation

Supplementary Readings

Links