metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: transcript
dtype: string
splits:
- name: train
num_bytes: 1166632952.762
num_examples: 36486
download_size: 1120249610
dataset_size: 1166632952.762
configs:
- config_name: default
data_files:
- split: train
path: data/train-*