metadata
dataset_info:
features:
- name: text
dtype: string
- name: audio
dtype: audio
- name: source
dtype: string
- name: text_revised
dtype: string
- name: language
dtype: string
splits:
- name: train
num_bytes: 21852630
num_examples: 63
download_size: 20891507
dataset_size: 21852630
configs:
- config_name: default
data_files:
- split: train
path: data/train-*