metadata
dataset_info:
features:
- name: base64
dtype: string
- name: transcript
dtype: string
splits:
- name: train
num_bytes: 1270627376
num_examples: 3500
download_size: 1186287686
dataset_size: 1270627376
configs:
- config_name: default
data_files:
- split: train
path: data/train-*