metadata
dataset_info:
features:
- name: audio
sequence: float32
- name: label
dtype: int64
splits:
- name: train
num_bytes: 333852048
num_examples: 922
download_size: 167550052
dataset_size: 333852048
configs:
- config_name: default
data_files:
- split: train
path: data/train-*