metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: metadata
sequence: string
splits:
- name: train
num_bytes: 455041001.86346364
num_examples: 1891
- name: test
num_bytes: 113820096.76353638
num_examples: 473
download_size: 558721591
dataset_size: 568861098.627
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*