metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: sentence
dtype: string
splits:
- name: train
num_bytes: 74289508175.27217
num_examples: 307740
- name: valid
num_bytes: 1159395389.6599138
num_examples: 4759
- name: test
num_bytes: 1139024019.0189137
num_examples: 4759
download_size: 75534073941
dataset_size: 76587927583.951
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*
- split: test
path: data/test-*