35: rCALMA Environment for Live Music Data Science

David M. Weigl, Kevin R. Page, Oxford e-Research Centre, University of Oxford

RCalma is a data science environment providing an analytical workflow for investigating large corpora of live music published by the Computational Analysis of the Live Music Archive (CALMA) project. RCalma aggregates and visualises Linked Data descriptions of concert metadata and audio-derived feature data published by CALMA. This allows us to answer questions of live music recordings beyond individual performances. We can visualise how the tempo of a song changes over time, as a band performs the song dozens or hundreds of times over a period of several years or decades; or, we can probe very high-level concepts, such as the typical distribution of musical key that a song is performed in over many renditions. In turn, this allows us to isolate and listen in to the most typical performances of a song, or indeed, the most deviant outliers, providing a means of discovery and data refinement.


Biographical notice:
David Weigl is a postdoctoral researcher in Music and Linked Data at the Oxford e-Research Centre. Lead developer of the Music Encoding and Linked Data (MELD) framework, and amateur funk bassist.

Kevin Page is a Senior researcher at the University of Oxford, leads the FAST Music Flows strand. His work applies Linked Data to the organisation and analysis of music and musical information.