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: 282251829.65
num_examples: 4670
- name: validation
num_bytes: 27293551
num_examples: 713
download_size: 215559289
dataset_size: 309545380.65
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*