metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
num_channels: 1
splits:
- name: train
num_bytes: 13094688031
num_examples: 24617
download_size: 10445221181
dataset_size: 13094688031
configs:
- config_name: default
data_files:
- split: train
path: data/train-*