Research project "The Jazzomat Research Project"

Representative: Deutsche Forschungsgemeinschaft (DFG)
Duration: 2012–2017

Project description

The main aim of the project is to investigate the creative processes underlying jazz solo improvisations with the help of statistical and computational methods. The the overarching goal is to explore the cognitive and cultural foundations of jazz solo improvisation. To this end, the Weimar Jazz Database, a comprehensive and representative database of jazz solo transcription is being built up, and an open-source Python Library, the MeloSpyLib is developed for analysis purposes. The project has a strong interdisciplinary approach situated at the intersection of jazz research, cognitive music psychology of creativity, and computational and statistical musicology, exploring research issues and methods of all these disciplines. Furthermore, the project has intersections with and implications for music information retrieval (MIR).

Our main research goals are:
  • to describe and discriminate different jazz improvisation styles by various features, e.g., with regard to the relations of pitch to metre, tonality and the underlying chords;
  • to compare features of jazz improvisation with features in other styles and areas of music;
  • to explore the cognitive foundations of improvisation while testing theories about the cognition of creative processes, e.g. by determining recurring melodic and rhythmic patterns and their underlying melodic-rhythmic prototypes;
  • the dependence of improvisational material on external parameters like instrument, tempo, key and their interrelationship;
  • to evaluate and enhance pedagogical approaches towards jazz improvisation;
  • to test and enhance statistical methods of music analysis, e.g., measures of complexity and coherence.



Research Fellows:

Prof. Dr. Martin Pfleiderer (Leitung)
Dr.-Ing. Jakob Abeßer
Dr. Klaus Frieler