VoxCeleb2

VoxCeleb2 is a dataset that contains a large collection of speech recordings from various speakers, used to evaluate the performance of voice-related tasks in machine learning and computer vision applications.

Papers using VoxCeleb2 (2)

Vidface: A Full-transformer Solver For Video Facehallucination With Unaligned Tiny Snapshots2021 · 2 citesRID-TWIN: An End-to-end Pipeline For Automatic Face De-identification In Videos2024