Pop music is one of the most popular forms of music around the world, enjoyed by millions of people every day. It is a cultural phenomenon that reflects the changing tastes, attitudes, and experiences of society, and has a profound impact on the way we see and understand the world around us. However, despite its popularity and significance, there is still much we do not know about Pop music as a cultural entity and about the social and cultural context in which it is produced and consumed.
To help answer such questions, we are using Music Information Retrieval (MIR) methods to study Western Pop music culture at a large scale. MIR is an interdisciplinary field that involves using advanced technology and computer methods to extract information from music, which includes signal processing, machine learning, and artificial intelligence. Our approach is reminiscent of "distant reading", which means using algorithmic analysis of thousands of books as a research tool in digital humanities. We will also use "distant listening" to audio using MIR methods, "distant viewing" of album artwork via object recognition, and "distant reading" of meta-information, lyrics, magazines, and books using natural language processing. These "distant" computer methods allow to study much greater numbers of songs, images and texts than individual experts are able to analyze. An integrative multi-modal approach is important because Pop music is not just about the music itself, but also about the cultural and social meanings that it carries.
Our main goal is to critically appraise MIR's ability to provide empirical evidence to existing Pop music theories from humanities. We will provide exemplary results concerning a number of questions in a data-driven manner. For example, can we chart the universe of topics in music lyrics? Can we compute an iconography of album cover art? Can we find evidence for retro cycles in Pop music's development? What is the influence of subcultures on mainstream music? By answering these questions, we hope to gain a deeper understanding of Pop music and its place in society. We will also need to be aware of the principal limitations of distant reading/listening/viewing and precisely research and document these boundaries.
This project will open up a new sub-field of MIR on the empirical and quantitative study of Pop music culture. This new sub-field will also help to establish a digital humanities approach to Pop music culture, which will help us to understand this cultural phenomenon in a more comprehensive way.
June 2023 - May 2026
Flexer A.: Can ChatGPT Be Useful for Distant Reading of Music Similarity?, opens an external URL in a new window, in HCMIR23: 2nd Workshop on Human-Centric Music Information Research, Milan, Italy, 2023.
Sturm B.L.T, Flexer A.: A Review of Validity and its Relationship to Music Information Research, opens an external URL in a new window, in Proc. of the 24th Int. Society for Music Information Retrieval Conf., Milan, Italy, 2023.