P. Cano
N. Wack
P. Herrera
2018-07-02
<p>This is a collection of audio used for the Genre Identification task of the ISMIR 2004 audio description contest organized by the Music Technology Group (Universitat Pompeu Fabra). The audio for the task was collected from <a href="http://magnatune.com/">Magnatune</a>, which contains a large amount of music licensed under Creative Commons licenses. The task of the contest was to classify a set of songs into genres, using the genre labels that Magnatune provided in their database.</p>
<p>Further information about the original contest and the contents of the dataset can be obtained from the following technical report:</p>
<blockquote>
<p>Cano P, Gómez E, Gouyon F, Herrera P, Koppenberger M, Ong B, Serra X, Streich S, Wack N. ISMIR 2004 audio description contest. Barcelona: Universitat Pompeu Fabra, Music technology Group; 2006. 20 p. Report No.: MTG-TR-2006-02</p>
</blockquote>
<p><a href="http://hdl.handle.net/10230/34013">http://hdl.handle.net/10230/34013</a></p>
<p>The original contest website can be found at <a href="http://ismir2004.ismir.net/genre_contest/">http://ismir2004.ismir.net/genre_contest/</a></p>
<p>The dataset contains the audio tracks from following 8 genres: classical, electronic, jazz- & blues, metal-, punk, rock-, pop, world.</p>
<p>For the genre recognition contest, the data was grouped into 6 classes: classical, electronic, jazz-blues, metal-punk, rock-pop, world, where in some cases two genres were merged into a single class. Note that ground-truth files uses these 6 classes, however in some cases the data is organised by original genre.</p>
<p><strong>Audio</strong></p>
<p>The audio is in MP3 format. It is divided into three folders, representing different subsets of the collection. Each folder has 729 files, split into classes. The number of files in each category reflects the proportion of files in each category in Magnatune when the dataset was created. No track appears in more than one folder.</p>
<ul>
<li>
<p><strong>Training:</strong> files for generating a classification model, arranged by class.</p>
</li>
<li>
<p><strong>Development:</strong> A separate set of files for participants to test their model against.</p>
</li>
<li>
<p><strong>Evaluation:</strong> originally a private subset, the files used to evaluate the accuracy of all submitted models</p>
</li>
</ul>
<p>The training and development set each consist of:</p>
<ul>
<li>
<p>classical: 320 files</p>
</li>
<li>
<p>electronic: 115 files</p>
</li>
<li>
<p>jazz_blues: 26 files</p>
</li>
<li>
<p>metal_punk: 45 files</p>
</li>
<li>
<p>rock_pop: 101 files</p>
</li>
<li>
<p>world: 122 files</p>
</li>
</ul>
<p>The evaluation set consists of 729 tracks with a similar distribution.</p>
<p><strong>Metadata</strong></p>
<p>Each folder of audio has a corresponding folder containing metadata of the files in that folder. The metadata is included in a file, tracklist.csv which has the following headers:</p>
<p>class, artist, album, track, track number, file path</p>
<p>The evaluation tracklist file has an additional column representing the magnatune track id of the recording.</p>
<p>Due to the way that the data was collected and distributed for the challenge, the metadata for the development subset is anonymised.</p>
<p><strong>Licensing</strong></p>
<p>The audio is licensed under a CC Attribution-NonCommercial-ShareAlike license (<a href="https://creativecommons.org/licenses/by-nc-sa/1.0/">https://creativecommons.org/licenses/by-nc-sa/1.0/</a>).</p>
<p><strong>Using this dataset</strong></p>
<p>We would highly appreciate if scientific publications of works partly based on this dataset cite the above publication.</p>
<p>We are interested in knowing if you find our datasets useful! If you use our dataset please email us at <a href="mailto:mtg-info@upf.edu">mtg-info@upf.edu</a> and tell us about your research.</p>
Licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 1.0 Generic license
https://doi.org/10.5281/zenodo.1302992
oai:zenodo.org:1302992
Zenodo
https://hdl.handle.net/10230/34013
https://zenodo.org/communities/mtgupf
https://zenodo.org/communities/mdm-dtic-upf
https://doi.org/10.5281/zenodo.1302991
info:eu-repo/semantics/openAccess
Other (Open)
ISMIR04 Genre Identification task dataset
info:eu-repo/semantics/other