metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: label
sequence: int64
splits:
- name: train
num_bytes: 2996659642.896
num_examples: 50224
- name: validation
num_bytes: 8447800428.846
num_examples: 31779
download_size: 11868081462
dataset_size: 11444460071.742
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*