metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: transcript
dtype: string
- name: duration
dtype: float64
splits:
- name: train
num_bytes: 1656118104.724
num_examples: 10372
- name: test
num_bytes: 709764768.105
num_examples: 4465
download_size: 2006614586
dataset_size: 2365882872.829
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*