metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: text
dtype: string
splits:
- name: train
num_bytes: 71082421385.436
num_examples: 16027
download_size: 64767811827
dataset_size: 71082421385.436
configs:
- config_name: default
data_files:
- split: train
path: data/train-*