metadata
dataset_info:
features:
- name: path
dtype: string
- name: audio
dtype: audio
- name: sentence
dtype: string
splits:
- name: train
num_bytes: 8034330005.745
num_examples: 28005
download_size: 6939942344
dataset_size: 8034330005.745
configs:
- config_name: default
data_files:
- split: train
path: data/train-*