metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: sentence
dtype: string
splits:
- name: train
num_bytes: 11501429490.519106
num_examples: 30859
- name: test
num_bytes: 2875455988.974702
num_examples: 7715
download_size: 13073083241
dataset_size: 14376885479.493809
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*