The Computational Music and Sound Archive project is a search engine for music which sorts musical pieces according to musical content, like rhythm, pitch, timbre or the like. Hidden-Markov Models (HMM) are used to estimate a fingerprint of the piece. Then these HMM are fed into a self-organizing Kohonen map which automatically sorts the pieces in a two-dimensional neuronal grid.
An International Symposium on Computational Ethnomusicological Archiving (ISCEA) was held in Dec. 2017 at the Institute of Systematic Musicology, University of Hamburg to discuss similar approaches and bring together ethnomusicologists, computer engineers, recording archives and museums.
The project is funded by the Deutsche Forschungsgemeinschaft (DFG).
The symposium is funded by the Volkswagen Stiftung.
Bader, R. (ed.): Computational Phonogram Archiving. Springer Series Current Research in Systematic Musicology. Springer, Heidelberg 2018 (in press).