metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: text
dtype: string
- name: start_time
dtype: string
- name: end_time
dtype: string
splits:
- name: train
num_bytes: 4763975
num_examples: 4
download_size: 4731364
dataset_size: 4763975
configs:
- config_name: default
data_files:
- split: train
path: data/train-*