---
dataset_info:
  features:
  - name: idx
    dtype: int64
  - name: audio
    dtype: audio
  - name: caption
    dtype: string
  splits:
  - name: train
    num_bytes: 5146352114.42
    num_examples: 26730
  download_size: 4177130146
  dataset_size: 5146352114.42
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---