README.md

metadata

dataset_info:
  features:
    - name: audio
      struct:
        - name: array
          sequence: float64
        - name: sampling_rate
          dtype: int64
    - name: speaker
      dtype: int64
  splits:
    - name: train
      num_bytes: 24572350028
      num_examples: 38435
    - name: test
      num_bytes: 1260853764
      num_examples: 2653
  download_size: 6421439087
  dataset_size: 25833203792
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*