Presentation Open Access

ISMIR 2019 tutorial: waveform-based music processing with deep learning

Jongpil Lee; Jordi Pons; Sander Dieleman

JSON-LD ( Export

  "description": "<p>A common practice when processing music signals with deep learning is to transform the raw waveform input into a time-frequency representation. This pre-processing step allows having less variable and more interpretable input signals. However, along that process, one can limit the model&#39;s learning capabilities since potentially useful information (like the phase or high frequencies) is discarded. In order to overcome the potential limitations associated with such pre-processing, researchers have been exploring waveform-level music processing techniques, and many advances have been made with the recent advent of deep learning.</p>\n\n<p>In this tutorial, we introduce three main research areas where waveform-based music processing can have a substantial impact:</p>\n\n<p>1) Classification: waveform-based music classifiers have the potential to simplify production and research pipelines.</p>\n\n<p>2) Source separation: making possible waveform-based music source separation would allow overcoming some historical challenges associated with discarding the phase.</p>\n\n<p>3) Generation: waveform-level music generation would enable, e.g., to directly synthesize expressive music.</p>\n\n<p><a href=\"\">Link to the original Google Slides</a></p>", 
  "license": "", 
  "creator": [
      "affiliation": "KAIST", 
      "@type": "Person", 
      "name": "Jongpil Lee"
      "affiliation": "Dolby Laboratories", 
      "@type": "Person", 
      "name": "Jordi Pons"
      "affiliation": "DeepMind", 
      "@type": "Person", 
      "name": "Sander Dieleman"
  "url": "", 
  "datePublished": "2019-11-04", 
  "@type": "PresentationDigitalDocument", 
  "@context": "", 
  "identifier": "", 
  "@id": "", 
  "workFeatured": {
    "url": "", 
    "alternateName": "ISMIR", 
    "location": "Delft", 
    "@type": "Event", 
    "name": "20th annual conference of the International Society for Music Information Retrieval"
  "name": "ISMIR 2019 tutorial: waveform-based music processing with deep learning"
All versions This version
Views 1,9501,944
Downloads 1,9791,978
Data volume 27.2 GB27.2 GB
Unique views 1,7161,710
Unique downloads 1,7321,731


Cite as