metadata
dataset_info:
features:
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 7969882825.3
num_examples: 18475
download_size: 6969873981
dataset_size: 7969882825.3
configs:
- config_name: default
data_files:
- split: train
path: data/train-*