metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: sentence
dtype: string
splits:
- name: train
num_bytes: 42675513028.354546
num_examples: 1308479
- name: test
num_bytes: 10668867764.897457
num_examples: 327120
download_size: 53288070178
dataset_size: 53344380793.252
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*