metadata
dataset_info:
features:
- name: file
dtype: audio
- name: text
dtype: string
- name: duration
dtype: float64
splits:
- name: train
num_bytes: 1442813569.665
num_examples: 12495
download_size: 1270558138
dataset_size: 1442813569.665
configs:
- config_name: default
data_files:
- split: train
path: data/train-*