metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: text
dtype: string
- name: source
dtype: string
splits:
- name: train
num_bytes: 7120511949.428776
num_examples: 14517
- name: too_long_test
num_bytes: 72593219.571224
num_examples: 148
download_size: 5917787275
dataset_size: 7193105169
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: too_long_test
path: data/too_long_test-*