Project "Jazzomat"

Funded by Deutsche Forschungsgemeinschaft (DFG), 10/2012-03/2017

The aim of the project is to explore the melodic-rhythmical structure of jazz solos. Approximately 250 jazz improvisations of various jazz styles and musicians will be transcribed semi-automatically and characterized using several musical features (e.g., pitch, interval, duration, metrical position) through the use of computational tools. The data will be analyzed in order to identify and classify melodic and rhythmical units (patterns and motifs). Occurrence and variation of patterns and motifs within a piece or across pieces of a particular jazz style or performer, will permit researchers to draw conclusions on creative processes and stylistic features. The results of the research project will offer insights for jazz research (jazz history, jazz theory) as well as for the psychology of musical creativity. The findings will also have implications for jazz education (didactics in improvisation) and music informatics (automatic transcription, statistical music analyzes, style classifications).


To the project's website