Skip to content

Dataset · ProgramComputer

VoxCeleb and VoxCeleb2 audio and video

ProgramComputer/voxceleb

Audio-video dataset for speech recognition and classification; combined multi-volume ZIP archives.

What's inside

A repository combining the VoxCeleb and VoxCeleb2 datasets. It contains audio and video files (mp4) packaged in multi-volume ZIP archives, with some files outside the original datasets. Metadata indicates a size between 100k and 1M examples. Rights remain with the original owners, and removal requests are possible.

What you can do with it

  • Training ASR on audio extracted from video
  • Speaker classification and verification
  • Training video and audio classifiers
Tasks
Automatic speech recognition, Audio classification, Image classification, Video classification
Size
100K–1M rows
Downloads this month
10.6K
Likes
130
Access
on request on Hugging Face

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("ProgramComputer/voxceleb")