metadata
dataset_info:
features:
- name: normalized_text
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 32099133623.71
num_examples: 95533
download_size: 29684723499
dataset_size: 32099133623.71
configs:
- config_name: default
data_files:
- split: train
path: data/train-*