README.md

metadata

dataset_info:
  features:
    - name: audio
      dtype:
        audio:
          sampling_rate: 16000
    - name: logits
      sequence: float32
  splits:
    - name: train
      num_bytes: 957958714.2162045
      num_examples: 29881
  download_size: 857931393
  dataset_size: 957958714.2162045
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*