Video/Audio Open Access
Delgado, Héctor; Evans, Nicholas; Kinnunen, Tomi; Lee, Kong Aik; Liu, Xuechen; Nautsch, Andreas; Patino, Jose; Sahidullah, Md; Todisco, Massimiliano; Wang, Xin; Yamagishi, Junichi
=== Notice ===
Evaluation keys (bona fide / spoof) and meta-labels can be downloaded from https://www.asvspoof.org/index2021.html.
The automatic speaker verification spoofing and countermeasures (ASVspoof) challenge series is a community-led initiative which aims to promote the consideration of spoofing and the development of countermeasures. ASVspoof 2021 is the 4th in a series of bi-annual, competitive challenges where the goal is to develop countermeasures capable of discriminating between bona fide and spoofed or deepfake speech.
ASVspoof 2021 comprises three different tasks:
Logical access (LA): bona fide and spoofed utterances generated using text-to-speech (TTS) and voice conversion (VC) algorithms are communicated across telephony and VoIP networks with various coding and transmission effects;
Physical access (PA): bona fide utterances are made in a real, physical space in which spoofing attacks are captured and then replayed within the same physical space using replay devices of varying quality;
Speech deepfake (DF): a fake audio detection task comprising bona fide and spoofed utterances generated using TTS and VC algorithms. Similar to the LA task (includes compressed data) but without speaker verification.
This repository provides the evaluation set for the DF task.
For more details, please visit our website below. https://www.asvspoof.org
|All versions||This version|
|Data volume||127.7 TB||127.7 TB|