metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: text
dtype: string
splits:
- name: train
num_bytes: 3590172528.232
num_examples: 23768
download_size: 3314950487
dataset_size: 3590172528.232
configs:
- config_name: default
data_files:
- split: train
path: data/train-*