metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: text
dtype: string
splits:
- name: train
num_bytes: 95257752243.132
num_examples: 541822
download_size: 82841027116
dataset_size: 95257752243.132
configs:
- config_name: default
data_files:
- split: train
path: data/train-*