Dataset Open Access

OpenMIC-2018

Humphrey, Eric J.; Durand, Simon; McFee, Brian


Citation Style Language JSON Export

{
  "publisher": "Zenodo", 
  "DOI": "10.5281/zenodo.1432913", 
  "title": "OpenMIC-2018", 
  "issued": {
    "date-parts": [
      [
        2018, 
        9, 
        23
      ]
    ]
  }, 
  "abstract": "<p>The OpenMIC-2018 dataset is made available through&nbsp;a collaboration between <a href=\"https://spotify.com/\">Spotify</a>&nbsp;and <a href=\"https://steinhardt.nyu.edu/marl/\">MARL@NYU</a>. Additionally, the cost of annotation was sponsored by Spotify, whose contributions to open-source research can be found online at the <a href=\"http://developer.spotify.com/\">developer site</a>, <a href=\"https://labs.spotify.com/\">engineering blog</a>, and <a href=\"https://spotify.github.io/\">public GitHub</a>.</p>\n\n<p>If you use this dataset, please cite the following work:</p>\n\n<blockquote>\n<p>Humphrey, Eric J., Durand, Simon, and McFee, Brian. &quot;OpenMIC-2018: An Open Dataset for Multiple Instrument Recognition.&quot; in Proceedings of the 19th International Society for Music Information Retrieval Conference (ISMIR), 2018. [<a href=\"https://bmcfee.github.io/papers/ismir2018_openmic.pdf\">pdf</a>]</p>\n</blockquote>\n\n<p>The dataset is made available by Spotify AB&nbsp;under a <a href=\"https://creativecommons.org/licenses/by/4.0/\">Creative Commons Attribution 4.0 International (CC BY 4.0)</a>&nbsp;license. The full terms of this license are included alongside this dataset.</p>\n\n<p>This dataset contains the following:</p>\n\n<ul>\n\t<li>10 second snippets of audio, in a directory format like &#39;audio/{0:3}/{0}.ogg&#39;.format(sample_key)</li>\n\t<li>VGGish features as JSON objects,&nbsp;in a directory format like &#39;vggish/{0:3}/{0}.json&#39;.format(sample_key)</li>\n\t<li>MD5 checksums for each OGG and JSON file</li>\n\t<li>Anonymized individual responses, in &#39;openmic-2018-individual-responses.csv&#39;</li>\n\t<li>Aggregated labels, in &#39;openmic-2018-aggregated-labels.csv&#39;</li>\n\t<li>Track metadata, with licenses for each audio recording, in &#39;openmic-2018-metadata.csv&#39;</li>\n\t<li>A Python-friendly NPZ file of features and labels, &#39;openmic-2018.npz&#39;</li>\n\t<li>Sample partitions for train and test, in &#39;partitions/*.txt&#39;</li>\n</ul>", 
  "author": [
    {
      "family": "Humphrey, Eric J."
    }, 
    {
      "family": "Durand, Simon"
    }, 
    {
      "family": "McFee, Brian"
    }
  ], 
  "version": "v1.0.0", 
  "type": "dataset", 
  "id": "1432913"
}
2,542
3,094
views
downloads
All versions This version
Views 2,5422,546
Downloads 3,0943,094
Data volume 8.1 TB8.1 TB
Unique views 2,0962,100
Unique downloads 1,1321,132

Share

Cite as