Daniel M├╝llensiefen

FANTASTIC: A Feature Analysis Toolbox for cognitive research on the perception of popular music (20/10/09)

FANTASTIC is an open source software toolbox for the analysis of melodies for research on music cognition and popular music. It operates by characterising melodies in terms of computational features. These features can be then employed to model human perception and cognition of melodies in lab studies. Alternatively, features might also be used to relate real-world musical phenomena to structural characteristics of tunes. The features included in FANTASTIC comprise summary features that reduce aspects of a tune to a numerical value and sequence-based features that are based on melodic sub-sequences (n-grams). FANTASTIC also implements second-order features which make use of the musical information contained in large corpora of music, thus they allow the researcher to consider different musical backgrounds and repertoires explicitly as part of a cognitive model. In addition to individual studies using this feature analysis approach and an overview of the general methodology, we will also give a very brief hands-on introduction on how to use the main functions of FANTASTIC for melodic feature analysis, corpus-based investigations and similarity computation.

Department of Computing, Goldsmiths College, University of London, New Cross, London, SE14 6NW

Tel: +44 (0) 20 7919 7850 | Fax: +44 (0) 20 7919 7853 | Email: computing@gold.ac.uk