metadata
dataset_info:
features:
- name: question
dtype: string
- name: answer
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
splits:
- name: 500k_part1_nonspeech
num_bytes: 4276857824
num_examples: 92413
- name: 500k_part1_speech
num_bytes: 3013971250
num_examples: 62965
- name: 500k_part2_nonspeech
num_bytes: 4347239182
num_examples: 93903
- name: 500k_part2_speech
num_bytes: 3048538557
num_examples: 63602
download_size: 14605005957
dataset_size: 14686606813
configs:
- config_name: default
data_files:
- split: 500k_part1_nonspeech
path: data/500k_part1_nonspeech-*
- split: 500k_part1_speech
path: data/500k_part1_speech-*
- split: 500k_part2_nonspeech
path: data/500k_part2_nonspeech-*
- split: 500k_part2_speech
path: data/500k_part2_speech-*