---
dataset_info:
  features:
  - name: text
    dtype: string
  - name: sample_rate
    dtype: int32
  - name: audio_tokens
    sequence: int64
  splits:
  - name: train
    num_bytes: 4556315682
    num_examples: 249269
  - name: validation
    num_bytes: 342638798
    num_examples: 18770
  download_size: 956277347
  dataset_size: 4898954480
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---