README.md

metadata

dataset_info:
  features:
    - name: audio
      dtype: audio
    - name: transcription
      dtype: string
  splits:
    - name: train
      num_bytes: 17159347574.893
      num_examples: 56427
  download_size: 11649243045
  dataset_size: 17159347574.893
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*