Voxceleb2

The Voxceleb2 dataset is a large-scale speaker recognition dataset, containing 2442 hours raw speech from 6112 speakers.

Data and Resources

Cite this as

J. S. Chung, A. Nagrani, A. Zisserman (2025). Dataset: Voxceleb2. https://doi.org/10.57702/otue2x0v

DOI retrieved: January 3, 2025

Additional Info

Field Value
Created January 3, 2025
Last update January 3, 2025
Defined In https://doi.org/10.48550/arXiv.2203.02195
Citation
  • https://doi.org/10.48550/arXiv.2304.13115
  • https://doi.org/10.48550/arXiv.2304.08945
  • https://doi.org/10.48550/arXiv.2203.04099
  • https://doi.org/10.48550/arXiv.2106.00273
Author J. S. Chung
More Authors
A. Nagrani
A. Zisserman
Homepage https://doi.org/10.1109/ICASSP.2018.8431638