metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: filename
dtype: string
splits:
- name: train
num_bytes: 19704770708
num_examples: 37552
download_size: 18410520344
dataset_size: 19704770708
configs:
- config_name: default
data_files:
- split: train
path: data/train-*