metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: sentence
dtype: string
- name: start_time
dtype: string
- name: end_time
dtype: string
splits:
- name: train
num_bytes: 30390526
num_examples: 108
- name: validation
num_bytes: 2207837
num_examples: 36
download_size: 8386474
dataset_size: 32598363
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*