Visualising Chord Progressions in Music Collections: A Big Data Approach

Kachkaev, A., Wolff, D., Barthet, M., Tidhar, D., Plumbley, M. D., Dykes, J. & Weyde, T. (2014). Visualising Chord Progressions in Music Collections: A Big Data Approach. Paper presented at the 9th Conference on Interdisciplinary Musicology – CIM14, 03-12-2014 - 06-12-2014, Staatliches Institut für Musikforschung, Berlin, Germany.

[img]
Preview
PDF - Published Version
Download (3MB) | Preview
Supplementary Materials:

Abstract

In the Digital Music Lab project we work on the automatic analysis of large audio databases that results in rich annotations for large corpora of music. The musicological interpretation of this data from thousands of pieces is a challenging task that can benefit greatly from specifically designed interactive visualisation. Most existing big music data visualisation focuses on cultural attributes, mood, or listener behaviour.

In this ongoing work we explore chord sequence patterns extracted by sequential pattern mining of more than one million tracks from the I Like Music commercial music collection. We present here several new visual representations that summarise chord patterns according to chord types, chroma, pattern structure and support, enabling musicologists to develop and answer questions about chord patterns in music collections.

Our visualisations represent root movement and chord qualities mostly in a geometrical way and use colour to represent pattern support. We use two individually configurable views in parallel to encourage comparisons, either between different representations of one corpus, highlighting complimentary musical aspects, or between different datasets,here representing different genres. We adapt several visualisation techniques to chord pattern sets using some novel layouts to support musicologists with their exploration and interpretation of the corpora. We found that differences between chord patterns of different genres, e.g. Rock & Roll vs. Jazz, are visible and can be used to generate hypotheses for the study of individual pieces, further statistical investigations or new data processing and visualisation. Our designs will be adapted as user needs are established through ongoing work. Means of aggregating, focusing and filtering by selected characteristics (such as key,melodic patterns etc.) will be added as we develop our design for the visualisation of chord patterns in close collaboration with musicologists.

Item Type: Conference or Workshop Item (Paper)
Subjects: M Music and Books on Music
Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: School of Informatics > Department of Computing
URI: http://openaccess.city.ac.uk/id/eprint/5629

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year

View more statistics