metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: text
dtype: string
- name: id
dtype: int64
splits:
- name: train
num_bytes: 3847211059.456
num_examples: 78089
download_size: 3156403433
dataset_size: 3847211059.456
configs:
- config_name: default
data_files:
- split: train
path: data/train-*