metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: label
dtype:
class_label:
names:
'0': media
'1': recordings
'2': unidentified
splits:
- name: train
num_bytes: 370981663502.674
num_examples: 719853
download_size: 552670139289
dataset_size: 370981663502.674