README.md

metadata

dataset_info:
  features:
    - name: audio
      struct:
        - name: array
          sequence: float64
        - name: sampling_rate
          dtype: int64
    - name: file
      dtype: string
    - name: speaker
      dtype: int64
  splits:
    - name: train
      num_bytes: 24577062310
      num_examples: 38435
  download_size: 6105986949
  dataset_size: 24577062310
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*