metadata
dataset_info:
features:
- name: captions
dtype: string
- name: location
dtype: audio
- name: duration
dtype: float64
splits:
- name: train
num_bytes: 541153267
num_examples: 434
download_size: 519501538
dataset_size: 541153267
configs:
- config_name: default
data_files:
- split: train
path: data/train-*