The MUSART Testbed for Query-By-Humming Evaluation

Show simple item record Roger Dannenberg en_US William Birmingham en_US George Tzanetakis en_US Colin Meek en_US Ning Hu en_US Bryan Pardo en_US
dc.contributor.editor Holger H. Hoos en_US
dc.contributor.editor David Bainbridge en_US 2004-10-21T04:26:22Z 2004-10-21T04:26:22Z 2003-10-26 en_US
dc.identifier.isbn 0-9746194-0-X en_US
dc.description.abstract Evaluating music information retrieval systems is acknowledged to be a difficult problem. We have created a database and a software testbed for the systematic evaluation of various query-by-humming (QBH) search systems. As might be expected, different queries and different databases lead to wide variations in observed search precision. "Natural" queries from two sources led to significantly lower performance than that typically reported in the QBH literature. These results point out the importance of careful measurement and objective comparisons to study retrieval algorithms. We compare string-matching, contour-matching, and hidden Markov model search algorithms in this study. An examination of scaling trends is encouraging: precision falls off very slowly as the database size increases. This trend is simple to compute and could be useful to predict performance on larger databases. en_US
dc.description.provenance Made available in DSpace on 2004-10-21T04:26:22Z (GMT). No. of bitstreams: 1 paper.pdf: 146236 bytes, checksum: 676e7f150999d7fa0e45e2f457e5fccb (MD5) Previous issue date: 2003-10-26 en
dc.format.extent 146236 bytes
dc.format.mimetype application/pdf
dc.language en en_US
dc.language.iso en_US
dc.publisher Johns Hopkins University en_US
dc.subject IR Systems and Algorithms en_US
dc.subject Digital Libraries en_US
dc.title The MUSART Testbed for Query-By-Humming Evaluation en_US
dc.type article en_US

Files in this item

Files Size Format Download
paper.pdf 146.2Kb application/pdf Download

This item appears in the following Collection(s)

Show simple item record