metadata
dataset_info:
features:
- name: text
dtype: string
- name: speech_tokens
list: int64
splits:
- name: train
num_bytes: 10085009
num_examples: 13100
download_size: 2763313
dataset_size: 10085009
configs:
- config_name: default
data_files:
- split: train
path: data/train-*