metadata
language:
- en
dataset_info:
features:
- name: url
dtype: string
- name: question
dtype: string
- name: answers
sequence: string
- name: audio
dtype:
audio:
sampling_rate: 16000
splits:
- name: train
num_bytes: 280021859.194
num_examples: 3778
- name: test
num_bytes: 152162547.968
num_examples: 2032
download_size: 407181214
dataset_size: 432184407.162
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*