metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 24000
- name: text
dtype: string
splits:
- name: train
num_bytes: 26836142243
num_examples: 145258
- name: test
num_bytes: 2980061780
num_examples: 16140
download_size: 26457456256
dataset_size: 29816204023
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*