Video/Audio Open Access

ASVspoof 2021 Challenge - Speech Deepfake Database

Delgado, Héctor; Evans, Nicholas; Kinnunen, Tomi; Lee, Kong Aik; Liu, Xuechen; Nautsch, Andreas; Patino, Jose; Sahidullah, Md; Todisco, Massimiliano; Wang, Xin; Yamagishi, Junichi


JSON Export

{
  "files": [
    {
      "links": {
        "self": "https://zenodo.org/api/files/5b5eff37-1f35-4b52-814d-ad05fd69b0cb/ASVspoof2021_DF_eval_part00.tar.gz"
      }, 
      "checksum": "md5:4f2cbae07cf3ede2a1dde3b8d2ee55ea", 
      "bucket": "5b5eff37-1f35-4b52-814d-ad05fd69b0cb", 
      "key": "ASVspoof2021_DF_eval_part00.tar.gz", 
      "type": "gz", 
      "size": 8637050238
    }, 
    {
      "links": {
        "self": "https://zenodo.org/api/files/5b5eff37-1f35-4b52-814d-ad05fd69b0cb/ASVspoof2021_DF_eval_part01.tar.gz"
      }, 
      "checksum": "md5:1578c89ab433c2b60b1dce93bdf8fbec", 
      "bucket": "5b5eff37-1f35-4b52-814d-ad05fd69b0cb", 
      "key": "ASVspoof2021_DF_eval_part01.tar.gz", 
      "type": "gz", 
      "size": 8624944890
    }, 
    {
      "links": {
        "self": "https://zenodo.org/api/files/5b5eff37-1f35-4b52-814d-ad05fd69b0cb/ASVspoof2021_DF_eval_part02.tar.gz"
      }, 
      "checksum": "md5:5497a35f0126e94a1d7a7d26db57b4f7", 
      "bucket": "5b5eff37-1f35-4b52-814d-ad05fd69b0cb", 
      "key": "ASVspoof2021_DF_eval_part02.tar.gz", 
      "type": "gz", 
      "size": 8616337195
    }, 
    {
      "links": {
        "self": "https://zenodo.org/api/files/5b5eff37-1f35-4b52-814d-ad05fd69b0cb/ASVspoof2021_DF_eval_part03.tar.gz"
      }, 
      "checksum": "md5:42b7512ba2943e98a32a53c9608cf83c", 
      "bucket": "5b5eff37-1f35-4b52-814d-ad05fd69b0cb", 
      "key": "ASVspoof2021_DF_eval_part03.tar.gz", 
      "type": "gz", 
      "size": 8656151261
    }
  ], 
  "owners": [
    226313
  ], 
  "doi": "10.5281/zenodo.4835108", 
  "stats": {
    "version_unique_downloads": 749.0, 
    "unique_views": 1128.0, 
    "views": 1245.0, 
    "version_views": 1245.0, 
    "unique_downloads": 749.0, 
    "version_unique_views": 1128.0, 
    "volume": 14270397995919.0, 
    "version_downloads": 1653.0, 
    "downloads": 1653.0, 
    "version_volume": 14270397995919.0
  }, 
  "links": {
    "doi": "https://doi.org/10.5281/zenodo.4835108", 
    "conceptdoi": "https://doi.org/10.5281/zenodo.4835107", 
    "bucket": "https://zenodo.org/api/files/5b5eff37-1f35-4b52-814d-ad05fd69b0cb", 
    "conceptbadge": "https://zenodo.org/badge/doi/10.5281/zenodo.4835107.svg", 
    "html": "https://zenodo.org/record/4835108", 
    "latest_html": "https://zenodo.org/record/4835108", 
    "badge": "https://zenodo.org/badge/doi/10.5281/zenodo.4835108.svg", 
    "latest": "https://zenodo.org/api/records/4835108"
  }, 
  "conceptdoi": "10.5281/zenodo.4835107", 
  "created": "2021-05-28T17:04:49.207732+00:00", 
  "updated": "2021-05-28T17:04:50.059629+00:00", 
  "conceptrecid": "4835107", 
  "revision": 1, 
  "id": 4835108, 
  "metadata": {
    "access_right_category": "success", 
    "doi": "10.5281/zenodo.4835108", 
    "version": "1.0", 
    "license": {
      "id": "ODbL-1.0"
    }, 
    "title": "ASVspoof 2021 Challenge - Speech Deepfake Database", 
    "related_identifiers": [
      {
        "scheme": "doi", 
        "identifier": "10.5281/zenodo.4835107", 
        "relation": "isVersionOf"
      }
    ], 
    "relations": {
      "version": [
        {
          "count": 1, 
          "index": 0, 
          "parent": {
            "pid_type": "recid", 
            "pid_value": "4835107"
          }, 
          "is_last": true, 
          "last_child": {
            "pid_type": "recid", 
            "pid_value": "4835108"
          }
        }
      ]
    }, 
    "grants": [
      {
        "code": "ANR-18-CE92-0024", 
        "links": {
          "self": "https://zenodo.org/api/grants/10.13039/501100001665::ANR-18-CE92-0024"
        }, 
        "title": "Reliable, secure and privacy preserving multi-biometric person authentication", 
        "acronym": "RESPECT", 
        "program": "", 
        "funder": {
          "doi": "10.13039/501100001665", 
          "acronyms": [
            "Not Available"
          ], 
          "name": "Agence Nationale de la Recherche", 
          "links": {
            "self": "https://zenodo.org/api/funders/10.13039/501100001665"
          }
        }
      }, 
      {
        "code": "309629", 
        "links": {
          "self": "https://zenodo.org/api/grants/10.13039/501100002341::309629"
        }, 
        "title": "NOTCH: NOn-cooperaTive speaker CHaracterization", 
        "acronym": "", 
        "program": "", 
        "funder": {
          "doi": "10.13039/501100002341", 
          "acronyms": [], 
          "name": "Academy of Finland", 
          "links": {
            "self": "https://zenodo.org/api/funders/10.13039/501100002341"
          }
        }
      }, 
      {
        "code": "ANR-18-JSTS-0001", 
        "links": {
          "self": "https://zenodo.org/api/grants/10.13039/501100001665::ANR-18-JSTS-0001"
        }, 
        "title": "Speaker identity cloning and protection", 
        "acronym": "VoicePersonae", 
        "program": "", 
        "funder": {
          "doi": "10.13039/501100001665", 
          "acronyms": [
            "Not Available"
          ], 
          "name": "Agence Nationale de la Recherche", 
          "links": {
            "self": "https://zenodo.org/api/funders/10.13039/501100001665"
          }
        }
      }
    ], 
    "keywords": [
      "Speaker recognition", 
      "Speaker verification", 
      "ASVspoof", 
      "Anti-spoofing", 
      "Presentation attack detection", 
      "Deepfake detection"
    ], 
    "publication_date": "2021-05-28", 
    "creators": [
      {
        "affiliation": "Nuance Communications Inc., Spain", 
        "name": "Delgado, H\u00e9ctor"
      }, 
      {
        "affiliation": "EURECOM, France", 
        "name": "Evans, Nicholas"
      }, 
      {
        "affiliation": "University of Eastern Finland, Finland", 
        "name": "Kinnunen, Tomi"
      }, 
      {
        "affiliation": "Institute for Infocomm Research, Singapore", 
        "name": "Lee, Kong Aik"
      }, 
      {
        "affiliation": "University of Eastern Finland, Finland", 
        "name": "Liu, Xuechen"
      }, 
      {
        "affiliation": "EURECOM, France", 
        "name": "Nautsch, Andreas"
      }, 
      {
        "affiliation": "EURECOM, France", 
        "name": "Patino, Jose"
      }, 
      {
        "affiliation": "Inria, France", 
        "name": "Sahidullah, Md"
      }, 
      {
        "affiliation": "EURECOM, France", 
        "name": "Todisco, Massimiliano"
      }, 
      {
        "affiliation": "National Institute of Informatics, Japan", 
        "name": "Wang, Xin"
      }, 
      {
        "affiliation": "National Institute of Informatics, Japan", 
        "name": "Yamagishi, Junichi"
      }
    ], 
    "access_right": "open", 
    "resource_type": {
      "type": "video", 
      "title": "Video/Audio"
    }, 
    "description": "<p>The automatic speaker verification spoofing and countermeasures (ASVspoof) challenge series is a&nbsp;community-led initiative which aims to promote the consideration of spoofing and the development of&nbsp;countermeasures. &nbsp;ASVspoof 2021 is the 4th in a series of bi-annual, competitive challenges where the&nbsp;goal is to develop countermeasures capable of discriminating between bona fide and spoofed or deepfake&nbsp;speech.&nbsp;<br>\n<br>\nASVspoof 2021 comprises three different tasks: &nbsp; &nbsp;<br>\n<em>&nbsp;Logical access (LA)</em>: bona fide and spoofed utterances generated using text-to-speech (TTS) and voice&nbsp;conversion (VC) algorithms are communicated across telephony and VoIP networks with various coding&nbsp;and transmission effects;<br>\n&nbsp; &nbsp;&nbsp;<br>\n<em>Physical access (PA)</em>: bona fide utterances are made in a real, physical space in which spoofing attacks&nbsp;are captured and then replayed within the same physical space using replay devices of varying quality;<br>\n&nbsp; &nbsp;&nbsp;<br>\n<em>Speech deepfake (DF)</em>: a fake audio detection task comprising bona fide and spoofed utterances&nbsp;generated using TTS and VC algorithms. Similar to the LA task (includes compressed data) but without&nbsp;speaker verification.<br>\n<br>\n<strong>This repository provides the evaluation set for the DF task.</strong><br>\n<br>\nFor more details, please visit our website below. https://www.asvspoof.org</p>"
  }
}
1,245
1,653
views
downloads
All versions This version
Views 1,2451,245
Downloads 1,6531,653
Data volume 14.3 TB14.3 TB
Unique views 1,1281,128
Unique downloads 749749

Share

Cite as