metadata
dataset_info:
features:
- name: base64
dtype: string
- name: transcript
dtype: string
splits:
- name: train
num_bytes: 1270665479
num_examples: 3500
download_size: 1186297241
dataset_size: 1270665479
configs:
- config_name: default
data_files:
- split: train
path: data/train-*