---
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
dataset_info:
  features:
  - name: audio_duration
    dtype: float64
  - name: audio_token
    sequence:
      sequence:
        sequence:
          sequence: int64
  - name: text
    dtype: string
  - name: text_token
    sequence: int64
  - name: align
    sequence: string
  splits:
  - name: train
    num_bytes: 94013092195
    num_examples: 2420047
  download_size: 16864044665
  dataset_size: 94013092195
---