We’re excited to announce the launch of Machine Learning for Music Information Retrieval, the second course in our Music Information Retrieval Program, developed by Professor George Tzanetakis at University of Victoria, Canada.

Want to know more about MIR? Read on below as Professor Tzanetakis explains.


Computers have profoundly influenced how music is created, distributed, and consumed. Music Information Retrieval (MIR) is the interdisciplinary science of retrieving information from music using computers. I started doing research in this field back in 1998, before the term MIR was used, and it has been the focus of my work since then. When I started working on algorithms to analyze music recordings many computers could not play sound and the hard disks of the time could only store a few audio files. Most listeners bought compact disks and heard music on the radio. Today, the large majority of listeners around the world experience music digitally using streaming music providers. These streaming services are powered by intelligent music recommendation algorithms that utilize MIR techniques to process and better understand millions of audio recordings. Being part of this transformation has been an incredible journey.

MIR techniques are also used in digital audio effects such as auto-tuning and harmonizing. Algorithms for beat tracking, pitch detection and time stretching are common features in digital audio editing and mixing software. Moreover, audio fingerprinting and watermarking techniques are used to digitally track and identify copyrighted content. Finally, MIR techniques are used by music creation systems such as DJ software, synthesizers, and music controllers.

In the last twenty years I have taught MIR topics to undergraduate and graduate students, presented tutorials to researchers in academia and industry, and supervised directed studies, as well as Masters and PhD theses on this topic. The highly interdisciplinary nature of this field is what attracted me to it but it is also what poses unique challenges when teaching. I am fortunate to have a solid formal music education in addition to my computer science training, but I can’t assume that’s the case for all the students who take my courses. Digital signal processing is a topic typically covered in Electrical and Computer Engineering courses but not in Computer Science. Over the years I am constantly trying to come up with more effective ways to teach all the different topics needed to understand MIR algorithms and techniques.

In 2016 I was approached by Kadenze to develop an online course for Music Information Retrieval. I was honored and excited about the possibility, as I was aware of their existing courses many of which were interdisciplinary and exploring creative technologies. MIR can be considered 20 years old and many of the research techniques that were initially explored in academic research are now an integral part of industrial applications. However, it is still an emerging field and there are many places around the world where there is no one available to teach it. As a young Computer Science undergraduate in Crete, Greece, I remember how thirsty I was for information about computer music as it was not covered in any courses I took and none of my professors had experience with it. To a large extent it was this desire to learn more about computers and music that motivated me to pursue graduate studies at Princeton University under the supervision of Perry Cook and led me to MIR. It is my hope that this Program will be a valuable resource for students and interested researchers from around the world and inspire them to do work in this field.

The Program covers the majority of topics that have been explored in the academic MIR community and provides a basic but solid introduction to concepts from digital signal processing, machine learning, and human-computer interaction from a MIR perspective. I have strived to make it accessible and useful to a variety of potential students ranging from motivated and curious first year students to experienced researchers in a variety of fields including Computer Science, Music, Electrical and Computer Engineering, Information Science, Musicology, and Psychology.

A course is like a garden. It takes a lot of preparation and work that is not immediately visible and requires regular monitoring and maintenance. In the next months and years I look forward to taking care of this Program and seeing how it evolves. I hope that taking this Program motivates you to learn more about MIR and undertake some work in this field. The annual conference in this field is ISMIR, the international conference of the Society for Music Information Retrieval and it is a great event. During ISMIR you might see an engineer explaining fourier transforms to a musicologist, or a music theorist explaining chord progression to a programmer. It is my hope that taking this Program will make you part of these conversations, and perhaps I’ll see you online or at the conference!

Discover the Program below:


To celebrate the launch, the Music Information Retrieval program is on sale for $USD350.00 – that’s $125 off the original price of $USD475.00. But only until August 1, 2020. Discover the program now and enroll to start learning!

LEARN MORE


About the Author and Program Instructor:
George Tzanetakis is a Professor in the Department of Computer Science with cross-listed appointments in ECE and Music at the University of Victoria, Canada. He is the Canada Research Chair (Tier II) in the Computer Analysis and Audio and Music, and received the Craigdaroch research award in artistic expression at the University of Victoria in 2012.  In 2011 he was Visiting Faculty at Google Research. He received his PhD in Computer Science at Princeton University in 2002, and was a Post-Doctoral fellow at Carnegie Mellon University in 2002-2003. His research spans all stages of audio content analysis such as feature extraction, segmentation, classification with specific emphasis on music information retrieval. He is also the primary designer and developer of Marsyas, an open source framework for audio processing with specific emphasis on music information retrieval applications. His pioneering work on musical genre classification received a IEEE signal processing society young author award and is frequently cited. More recently, he has been exploring new interfaces for musical expression, music robotics, computational ethnomusicology, and computer-assisted music instrument tutoring.