metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: sentence
dtype: string
splits:
- name: train
num_bytes: 441186222
num_examples: 20000
download_size: 496914511
dataset_size: 441186222
configs:
- config_name: default
data_files:
- split: train
path: data/train-*