Course syllabus — AY2024

Please read the course policies carefully, prior to taking this course.

Course overview

This course introduces and explores computational musicology, with a focus on concepts and techniques from the contemporary field of music information retrieval (MIR), and is aimed at students who have an interest in both music and technology. Students will learn to apply some of the basics of MIR to analyse music through various machine learning techniques.

This course will be delivered through a series of interactive lectures and workshops. There will be two projects, and quizzes, to test your knowledge of the course material. Feedback on assignment work will be given within approximately two weeks, via Manaba. You will also be assessed on your engagement in class activities.

Prerequisites

While there is no official prerequisite course for this class, it is strongly advised that only students who have taken the Introduction to Algorithms and Programming class, or have equivalent programming experience, take this class.

No prior formal study of music is necessary, but it is assumed that students who take this course have at least some basic knowledge of music, such as tones, chords, scales, rhythm, melody, harmony, and so on.

Note: Students who wish to take this course should ensure that they can get Python running on their computer!

Course objectives

Students who successfully complete this course will be able to:

• Understand and describe what music information retrieval is, and its applications to audio signals.
• Understand, know and apply (some of) the techniques of music information retrieval to extract and analyse the musical elements and qualities of digital audio content using contemporary software techniques.
• Understand and describe how artificial intelligence software can be used to understand and analyse audio.

Course readings

Selected readings from:

• Schafer, C. (2021). Quickstart Python: An introduction to programming for STEM students. Springer. https://doi.org/10.1007/978-3-658-33552-6 — Especially Chapters 1-9
• Schedl, M., Gómez, E., & Urbano, J. (2014). Music Information Retrieval: Recent developments and applications. Foundations and Trends in Information Retrieval, 8(2-3), 127–261. https://julian-urbano.info/files/publications/059-music-information-retrieval-recent-developments-applications.pdf — Especially Chapters 1-3

There is no need to buy physical copies of these sources as they are available digitally for free via the above URLs. The following book, available digitally via the Ritsumeikan VPN, will also be a useful (though quite technical) resource:

• Müller, M. (2015). Fundamentals of Music Processing: Audio, Analysis, Algorithms, Applications. Springer. https://link.springer.com/book/10.1007/978-3-319-21945-5

The website musicinformationretrieval.com also contains useful information and code examples that will be utilised in this class.

Software

In addition to Python 3, we will mainly rely on the Librosa package for the MIR part of the course, but will also cover other useful libraries for MIR, signal processing, and data management.

Weekly schedule

1. Introduction (Reading chapter 1 of Schedl et al)
2. Data types and data structures in Python (Reading chapters 2-5 of Schafer)
3. Flow controls (Reading chapter 6 of Schafer)
4. Functions, useful Python libraries (Reading chapters 7-9 of Schafer)
5. In-class workshop: Python Basics 
6. Audio formats (Reading chapter 3 of Schedl et al) — sample rate and bit depth; common audio formats, wav, mp3, etc.
7. Time-frequency representations — Spectrograms
8. Onset and novelty detection (Reading chapter 2 of Schedl et al)
9. In-class workshop: Temporal features — Assignment 1 due 5pm Wednesday November 27 
10. Pitch and Chroma
11. Musical key
12. Musical key (continued)
13. Musical similarity; music information retrieval (MIR) tasks
14. In-class workshop: MIR
15. In-class quiz — Assignment 2 due 9am the day of the final class

Assessments

• Assignments: 60% (30% x 2)
• Quiz: 15%
• Class engagement: 25%

Further details of assessments, including due dates, will be announced in class. Feedback for all assessment items will be given through Manaba and, where appropriate, in-person. Feedback on assignments will generally be returned within two weeks (10 working days), unless notified otherwise.

Assignments will cover one and two of the course objectives, while the quiz will cover one and three of the course objectives. Engagement in class activities is relevant to all of the course objectives.

Assignments

There will be two programming-focused assignments, based on the course content.

Quiz

At least one multiple-choice and short-answer quiz/test will be delivered throughout the semester. Questions will be directly related to the course content and readings.

Engagement in class activities

Your engagement will be based on the following criteria, with each criterion having equal weight:

• The student displays a positive attitude towards their own learning.
• The student is focused on and engaged in tasks and activities in class.
• The student proactively contributes to class by offering ideas and asking questions.
• The student comes to class prepared, having done required reading or work beforehand.
• The student is prompt and attends classes.

Note that any absences unaccounted for will affect this grade. Being late by 10 minutes, without a compelling reason, will count as an absence.

Citation style

All written work should use APA 7 referencing format. If you are modifying code snippets that you've found, you should include a reference in your code comments of where you found the code. You can find a "quickstart" guide at http://www.ritsumei.ac.jp/~haimes/publications/citation.pdf. A more comprehensive resource is the official APA 7 guide at https://apastyle.apa.org.

Back to top | © Paul Haimes at Ritsumeikan University