metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: transcript
dtype: string
splits:
- name: train
num_bytes: 9848101426.557
num_examples: 67189
download_size: 10348315897
dataset_size: 9848101426.557
configs:
- config_name: default
data_files:
- split: train
path: data/train-*