metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: labels
dtype: string
splits:
- name: train
num_bytes: 2711834755.57
num_examples: 20914
download_size: 2707887140
dataset_size: 2711834755.57
configs:
- config_name: default
data_files:
- split: train
path: data/train-*