metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: transcript
dtype: string
splits:
- name: train
num_bytes: 3526492077.614348
num_examples: 5346
- name: validation
num_bytes: 392492117.9476519
num_examples: 595
download_size: 3035853469
dataset_size: 3918984195.562
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*