metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: text
dtype: string
- name: id
dtype: int64
splits:
- name: train
num_bytes: 1214077398.792
num_examples: 27448
download_size: 1322317794
dataset_size: 1214077398.792
configs:
- config_name: default
data_files:
- split: train
path: data/train-*