Demo: Music Search Based on Aesthetic Similarity



This page contains search results generated by a music search engine prototype. The system uses power-law metrics to extract statistical proportions of music-theoretic and other attributes of music, as per:

1. B. Manaris, J. Romero, P. Machado, D. Krehbiel, T. Hirzel, W. Pharr, and R.B. Davis, "Zipf's Law, Music Classification and Aesthetics," Computer Music Journal 29(1), MIT Press, pp. 55-69, Spring 2005.

2. P. Roos and B. Manaris, "A Music Information Retrieval Approach Based on Power Laws," Proceedings of 19th IEEE International Conference on Tools with Artificial Intelligence (ICTAI-07), Patras, Greece, Oct. 2007, (to appear).

Usage: Given a source piece, the search engine outputs the three most aesthetically similar pieces from a corpus of 15,200+ MIDI files (e.g., Baroque, Classical, Romantic, Impressionist, Modern, Jazz, Rock, Country, and Pop).


Source Piece: Jazz, Miles Davis: Blue in Green
Search Results (ordered by MSE):


Source Piece: PALESTRINA, Giovanni Pierluigi da (c.1525-94): Hodie Christus natus est (B.F.M.Geurts)
Search Results (ordered by MSE):


Source Piece: BEETHOVEN, Ludwig van (1770-1827): 8 Lieder, Op.52, 6.Das Blümchen Wunderhold (S.S.Eskandari)
Search Results (ordered by MSE):


Source Piece: Jazz, Thelonious Monk: Blue Monk
Search Results (ordered by MSE):


Source Piece: BACH, Johann Sebastian (1685-1750): Two-Part Invention No.13 in A-, BWV784 (Cakewalk demo)
Search Results (ordered by MSE):