metadata
dataset_info:
features:
- name: filename
dtype:
audio:
sampling_rate: 16000
- name: 'Y'
dtype: string
splits:
- name: train
num_bytes: 130227530
num_examples: 4000
download_size: 130068678
dataset_size: 130227530
configs:
- config_name: default
data_files:
- split: train
path: data/train-*