metadata
dataset_info:
features:
- name: filename
dtype:
audio:
sampling_rate: 16000
- name: 'Y'
dtype: string
splits:
- name: train
num_bytes: 550593113
num_examples: 4500
download_size: 549213849
dataset_size: 550593113
configs:
- config_name: default
data_files:
- split: train
path: data/train-*