README.md

metadata

dataset_info:
  features:
    - name: id
      dtype: string
    - name: audio
      dtype: audio
    - name: sentence
      dtype: string
  splits:
    - name: train
      num_bytes: 12400028580.739
      num_examples: 100767
    - name: validation
      num_bytes: 519447209.472
      num_examples: 3783
    - name: test
      num_bytes: 596361411.064
      num_examples: 3837
  download_size: 11656859636
  dataset_size: 13515837201.275
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: validation
        path: data/validation-*
      - split: test
        path: data/test-*