metadata
dataset_info:
features:
- name: speech_tokens
dtype: string
- name: text
dtype: string
splits:
- name: train
num_bytes: 438902817
num_examples: 358503
download_size: 234126629
dataset_size: 438902817
configs:
- config_name: default
data_files:
- split: train
path: data/train-*