Thesis Open Access

From heuristics-based to data-driven audio melody extraction

Bosch, Juan J.


JSON-LD (schema.org) Export

{
  "inLanguage": {
    "alternateName": "eng", 
    "@type": "Language", 
    "name": "English"
  }, 
  "description": "<p><strong>Abstract</strong></p>\n\n<p>The identification of the melody from a music recording is a relatively easy task for humans, but very challenging for computational systems. This task is known as &quot;audio melody extraction&quot;, more formally defined as the automatic estimation of the pitch sequence of the melody directly from the audio signal of a polyphonic music recording. This thesis investigates the benefits of exploiting knowledge automatically derived from data for audio melody extraction, by combining digital signal&nbsp;processing and machine learning methods. We extend the scope of melody extraction research by working with a varied dataset and multiple definitions of melody. We first present an overview of the state of the art, and perform an evaluation focused on a novel symphonic music dataset. We then propose melody extraction methods based on a source-filter model and pitch contour characterisation and evaluate them on a wide range of music genres. Finally, we explore novel timbre, tonal and spatial features for contour characterisation, and propose a method for estimating multiple melodic lines. The combination of supervised and unsupervised approaches leads to advancements on melody extraction and shows a promising path for future research and applications.</p>\n\n<p>&nbsp;</p>\n\n<p><strong>Datasets:&nbsp;</strong><br>\n<br>\nThe symphonic music dataset proposed in this thesis (Orchset) is available at:</p>\n\n<p><a href=\"https://zenodo.org/record/1289786#.XnNV15P0mL8\">https://zenodo.org/record/1289786#.XnNV15P0mL8</a></p>\n\n<p>Orchset is intended to be used as a dataset for the development and evaluation of melody extraction algorithms. This collection contains 64 audio excerpts focused on symphonic music. with their corresponding annotation of the melody.</p>\n\n<p><strong>Code:</strong></p>\n\n<p>The source code of the melody extraction algorithms proposed in this thesis is available at:</p>\n\n<p><a href=\"https://github.com/juanjobosch/SourceFilterContoursMelody\">https://github.com/juanjobosch/SourceFilterContoursMelody</a></p>", 
  "license": "https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode", 
  "creator": [
    {
      "affiliation": "Universitat Pompeu Fabra, Barcelona", 
      "@id": "https://orcid.org/0000-0003-4221-3517", 
      "@type": "Person", 
      "name": "Bosch, Juan J."
    }
  ], 
  "sameAs": [
    "http://mtg.upf.edu/node/3737"
  ], 
  "image": "https://zenodo.org/static/img/logos/zenodo-gradient-round.svg", 
  "datePublished": "2017-06-27", 
  "headline": "From heuristics-based to data-driven audio melody extraction", 
  "url": "https://zenodo.org/record/1120334", 
  "keywords": [
    "Melody Extraction", 
    "Automatic", 
    "MIR", 
    "Music", 
    "Retrieval", 
    "Symphonic", 
    "Instrument", 
    "Agreement", 
    "Tonality", 
    "Timbre", 
    "Stereo", 
    "Source-filter", 
    "Separation", 
    "NMF", 
    "Visualisation", 
    "Evaluation", 
    "Dataset", 
    "Contour", 
    "Salience", 
    "Pitch", 
    "Supervised"
  ], 
  "@context": "https://schema.org/", 
  "identifier": "https://doi.org/10.5281/zenodo.1120334", 
  "@id": "https://doi.org/10.5281/zenodo.1120334", 
  "@type": "ScholarlyArticle", 
  "name": "From heuristics-based to data-driven audio melody extraction"
}
128
124
views
downloads
All versions This version
Views 128128
Downloads 124124
Data volume 1.3 GB1.3 GB
Unique views 107107
Unique downloads 111111

Share

Cite as