metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: text
dtype: string
- name: pinyin
dtype: string
splits:
- name: train
num_bytes: 22036529551.161663
num_examples: 63262
- name: test
num_bytes: 7039602241.368337
num_examples: 24773
download_size: 24652938931
dataset_size: 29076131792.53
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*