metadata
dataset_info:
features:
- name: url
dtype: string
- name: question
dtype: string
- name: answers
sequence: string
- name: question_unit
sequence: int64
- name: response_interleaf
dtype: string
- name: response_text
dtype: string
- name: response_tokens
sequence: int64
- name: response_speech
dtype: audio
splits:
- name: test
num_bytes: 1743859705
num_examples: 2032
download_size: 1677606128
dataset_size: 1743859705
configs:
- config_name: default
data_files:
- split: test
path: data/test-*