---
dataset_info:
  features:
  - name: caption
    dtype: string
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
  splits:
  - name: test
    num_bytes: 4897858523.041246
    num_examples: 15463
  - name: train
    num_bytes: 58584881989.45539
    num_examples: 183631
  download_size: 61916988256
  dataset_size: 63482740512.496635
configs:
- config_name: default
  data_files:
  - split: test
    path: data/test-*
  - split: train
    path: data/train-*
---