---
dataset_info:
  features:
  - name: docid
    dtype: string
  - name: audio
    dtype: audio
  splits:
  - name: train
    num_bytes: 47478046461.732
    num_examples: 51812
  download_size: 43932097937
  dataset_size: 47478046461.732
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---