Thesis Open Access

SATB Voice Segregation For Monoaural Recordings

Pétermann, Darius A,


JSON Export

{
  "files": [
    {
      "links": {
        "self": "https://zenodo.org/api/files/a587860c-c072-4414-be8a-92017192e58d/2020-Darius-Petermann.pdf"
      }, 
      "checksum": "md5:0d052cb5f9231ccff0caf7a2ffc406c4", 
      "bucket": "a587860c-c072-4414-be8a-92017192e58d", 
      "key": "2020-Darius-Petermann.pdf", 
      "type": "pdf", 
      "size": 12076163
    }
  ], 
  "owners": [
    46759
  ], 
  "doi": "10.5281/zenodo.4091247", 
  "stats": {
    "version_unique_downloads": 130.0, 
    "unique_views": 187.0, 
    "views": 221.0, 
    "version_views": 221.0, 
    "unique_downloads": 130.0, 
    "version_unique_views": 187.0, 
    "volume": 1847652939.0, 
    "version_downloads": 153.0, 
    "downloads": 153.0, 
    "version_volume": 1847652939.0
  }, 
  "links": {
    "doi": "https://doi.org/10.5281/zenodo.4091247", 
    "conceptdoi": "https://doi.org/10.5281/zenodo.4091246", 
    "bucket": "https://zenodo.org/api/files/a587860c-c072-4414-be8a-92017192e58d", 
    "conceptbadge": "https://zenodo.org/badge/doi/10.5281/zenodo.4091246.svg", 
    "html": "https://zenodo.org/record/4091247", 
    "latest_html": "https://zenodo.org/record/4091247", 
    "badge": "https://zenodo.org/badge/doi/10.5281/zenodo.4091247.svg", 
    "latest": "https://zenodo.org/api/records/4091247"
  }, 
  "conceptdoi": "10.5281/zenodo.4091246", 
  "created": "2020-10-15T14:27:48.739042+00:00", 
  "updated": "2020-10-16T00:26:56.812983+00:00", 
  "conceptrecid": "4091246", 
  "revision": 4, 
  "id": 4091247, 
  "metadata": {
    "access_right_category": "success", 
    "doi": "10.5281/zenodo.4091247", 
    "description": "<p>Choral singing is a widely practiced form of ensemble singing wherein a group of people sing simultaneously in polyphonic harmony. The most commonly practiced&nbsp;setting for choir ensembles consists of four parts; Soprano, Alto, Tenor and Bass&nbsp;(SATB), each with its own range of fundamental frequencies (F0s). The task of&nbsp;source separation for this choral setting entails separating the SATB mixture into&nbsp;its constituent parts. Source separation for musical mixtures is well studied and&nbsp;many Deep Learning-based methodologies have been proposed for the same. However,<br>\nmost of the research has been focused on a typical case which consists in<br>\nseparating vocal, percussion and bass sources from a mixture, each of which has a&nbsp;distinct spectral structure. In contrast, the simultaneous and harmonic nature of&nbsp;ensemble singing leads to high structural similarity and overlap between the spectral&nbsp;components of the sources in a choral mixture, making source separation for&nbsp;choirs a harder task than the typical case. This, along with the lack of an appropriate&nbsp;consolidated dataset has led to a dearth of research in the field so far. In&nbsp;this work we first assess how well some of the recently developed methodologies for&nbsp;musical source separation perform for the case of SATB choirs. We then propose a&nbsp;novel domain-specific adaptation for conditioning the recently proposed U-Net architecture<br>\nfor musical source separation using the fundamental frequency contour of<br>\neach of the singing groups and demonstrate that our proposed approach surpasses&nbsp;results from domain-agnostic architectures. Lastly we assess our approach using&nbsp;different evaluation methodologies, going from objective to subjective-based ones,&nbsp;and provide a comparative analysis of the various results.</p>", 
    "license": {
      "id": "CC-BY-3.0"
    }, 
    "title": "SATB Voice Segregation For Monoaural Recordings", 
    "relations": {
      "version": [
        {
          "count": 1, 
          "index": 0, 
          "parent": {
            "pid_type": "recid", 
            "pid_value": "4091246"
          }, 
          "is_last": true, 
          "last_child": {
            "pid_type": "recid", 
            "pid_value": "4091247"
          }
        }
      ]
    }, 
    "communities": [
      {
        "id": "mtgupf"
      }, 
      {
        "id": "smc-master"
      }
    ], 
    "thesis": {
      "university": "Universitat Pompeu Fabra", 
      "supervisors": [
        {
          "affiliation": "Universitat Pompeu Fabra", 
          "name": "Chandna, Pritish"
        }, 
        {
          "affiliation": "Universitat Pompeu Fabra", 
          "name": "Bonada, Jordi"
        }
      ]
    }, 
    "keywords": [
      "source separation, singing voice, SATB recording, convolutional neural networks, choir music."
    ], 
    "publication_date": "2020-09-15", 
    "creators": [
      {
        "affiliation": "Universitat Pompeu Fabra", 
        "name": "P\u00e9termann, Darius A,"
      }
    ], 
    "access_right": "open", 
    "resource_type": {
      "subtype": "thesis", 
      "type": "publication", 
      "title": "Thesis"
    }, 
    "related_identifiers": [
      {
        "scheme": "doi", 
        "identifier": "10.5281/zenodo.4091246", 
        "relation": "isVersionOf"
      }
    ]
  }
}
221
153
views
downloads
All versions This version
Views 221221
Downloads 153153
Data volume 1.8 GB1.8 GB
Unique views 187187
Unique downloads 130130

Share

Cite as