metadata
dataset_info:
features:
- name: image
dtype: image
- name: audio_file
dtype: string
- name: slice
dtype: int16
splits:
- name: train
num_bytes: 27839749
num_examples: 13040
download_size: 29038309
dataset_size: 27839749
configs:
- config_name: default
data_files:
- split: train
path: data/train-*