README.md

metadata

dataset_info:
  features:
    - name: path_to_audio
      dtype:
        audio:
          sampling_rate: 16000
    - name: abuse
      dtype: string
    - name: language
      dtype: string
  splits:
    - name: train
      num_bytes: 1597519981.996
      num_examples: 2458
    - name: test
      num_bytes: 717951151.872
      num_examples: 1112
  download_size: 1689763072
  dataset_size: 2315471133.868
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*