dc.contributor.author | STURM, BLT | en_US |
dc.contributor.author | Kereliuk, C | en_US |
dc.contributor.author | Larsen, J | en_US |
dc.contributor.author | Mathematics and Computation in Music | en_US |
dc.date.accessioned | 2016-04-22T14:03:35Z | |
dc.date.accessioned | 2016-10-27T15:26:36Z | |
dc.identifier.uri | http://qmro.qmul.ac.uk/xmlui/handle/123456789/16136 | |
dc.description.abstract | The ``winning'' system in the 2013 MIREX Latin Genre Classification Task was a deep neural network trained with simple features. An explanation for its winning performance has yet to be found. In previous work, we built similar systems using the {\em BALLROOM} music dataset, and found their performances to be greatly affected by slightly changing the tempo of the music of a test recording. In the MIREX task, however, systems are trained and tested using the {\em Latin Music Dataset (LMD)}, which is 4.5 times larger than {\em BALLROOM}, and which does not seem to show as strong a relationship between tempo and label as {\em BALLROOM}. In this paper, we reproduce the ``winning'' deep learning system using {\em LMD}, and measure the effects of time dilation on its performance. We find that tempo changes of at most $\pm 6$\% greatly diminish and improve its performance. Interpreted with the low-level nature of the input features, this supports the conclusion that the system is exploiting some low-level absolute time characteristics to reproduce ground truth in {\em LMD}. | en_US |
dc.language.iso | en | en_US |
dc.relation.replaces | http://qmro.qmul.ac.uk/xmlui/handle/123456789/12011 | |
dc.relation.replaces | 123456789/12011 | |
dc.title | ?`El Caballo Viejo? Latin Genre Recognition with Deep Learning and Spectral Periodicity | en_US |
dc.type | Conference Proceeding | |
dc.rights.holder | © 2015, Springer International Publishing Switzerland | |
dc.identifier.doi | 10.1007/978-3-319-20603-5_34 | |
pubs.notes | Not known | en_US |