---
dataset_info:
  features:
  - name: audio_token
    sequence:
      sequence:
        sequence: int64
  - name: text
    dtype: string
  - name: speaker
    dtype: string
  - name: book
    dtype: string
  - name: text_token
    sequence: int64
  splits:
  - name: train
    num_bytes: 5458701902
    num_examples: 205035
  download_size: 860267374
  dataset_size: 5458701902
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---