metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: label
dtype: int64
splits:
- name: train
num_bytes: 1990994168.312
num_examples: 6256
download_size: 1924390785
dataset_size: 1990994168.312
configs:
- config_name: default
data_files:
- split: train
path: data/train-*