---
dataset_info:
  features:
  - name: write format
    dtype: string
  - name: speak format
    dtype: string
  splits:
  - name: train
    num_bytes: 13223140117
    num_examples: 16043579
  download_size: 7412197927
  dataset_size: 13223140117
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---