README.md

metadata

dataset_info:
  features:
    - name: audio
      dtype: audio
    - name: text
      dtype: string
    - name: source
      dtype: string
    - name: sample_rate
      dtype: int64
    - name: speaker
      dtype: string
  splits:
    - name: train
      num_bytes: 20328604229
      num_examples: 30606
  download_size: 20684866085
  dataset_size: 20328604229
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*