Datasets:

confit
/

audioset-16khz-wds

configs:
  - config_name: 20k
    data_files:
      - split: train
        path: 20k/train/*.tar
      - split: test
        path: 20k/test/*.tar
language:
  - en
task_categories:
  - audio-classification
tags:
  - audio
size_categories:
  - 1M<n<10M