Musical style identification using self-organising maps

In this paper the capability of using self-organising neural maps (SOM) as music style classifiers from symbolic specifications of musical fragments is studied. From MIDI file sources, the monophonic melody track is extracted and cut into fragments of equal length. From these sequences, melodic, harmonic, and rhythmic numerical descriptors are computed and presented to the SOM. Their performance is analysed in terms of separability in different music classes from the activations of the map, obtaining different degrees of success for classical and jazz music. This scheme has a number of applications like indexing and selecting musical databases or the evaluation of style-specific automatic composition systems.

[1]  Otávio Augusto S. Carpinteiro.  A self-organizing map model for analysis of musical time series , 1998, Proceedings 5th Brazilian Symposium on Neural Networks (Cat. No.98EX209).

[2]  Hagen Soltau,et al.  Recognition of music types , 1998, Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '98 (Cat. No.98CH36181).

[3]  Andreas Rauber,et al.  Automatically Analyzing and Organizing Music Archives , 2001, ECDL.

[4]  Teuvo Kohonen,et al.  The self-organizing map , 1990 .

[5]  Steve Lawrence,et al.  Artist detection in music with Minnowmatch , 2001, Neural Networks for Signal Processing XI: Proceedings of the 2001 IEEE Signal Processing Society Workshop (IEEE Cat. No.01TH8584).

[6]  Jorma Laaksonen,et al.  SOM_PAK: The Self-Organizing Map Program Package , 1996 .

[7]  Belinda Thom,et al.  Unsupervised Learning and Interactive Jazz/Blues Improvisation , 2000, AAAI/IAAI.