metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: text
dtype: string
- name: duration
dtype: float32
splits:
- name: train
num_bytes: 19007240
num_examples: 14
- name: eval
num_bytes: 2288342
num_examples: 3
download_size: 21300610
dataset_size: 21295582
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: eval
path: data/eval-*