---
dataset_info:
  features:
  - name: text
    dtype: string
  - name: audio_token
    sequence:
      sequence:
        sequence:
          sequence: int64
  - name: idx
    dtype: int64
  splits:
  - name: train
    num_bytes: 87483072224
    num_examples: 8282988
  download_size: 14081502127
  dataset_size: 87483072224
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---