The music is increasingly predictable

Jul 27, 2012

The results indicate that the music tends to be more and more homogeneous and less sounds

​Thanks to the digitization of Western music we can use computers to listen and analyze music differently than humans do. The ability to analyze large-scale musical parameters with computers is a breakthrough that has allowed us to understand a different one of the more socially valued cultural fields, the songs. This is shown in a study involving researchers from the research group in Music Technology and published July 26 in Scientific Reports, a publication of the Nature publishing group. Alvaro Corral, is co-author and IP Group of Complex Systems Mathematical Research Center. Juan Serrano, lead author of the work, is a doctor by the UPF and is currently on a postdoctoral stay at the Institute of Artificial Intelligence Research, CSIC. The research involved the study of large-scale evolution of the basic statistical parameters of the audio information in a half-million songs from Western popular music between 1955 and 2010. The most relevant result of this study was that for the first time it has been proven empirically, objective and quantifiable certain unknown trends in music production as well as some unpublished data on the evolution of contemporary popular music.