metadata
dataset_info:
features:
- name: text
dtype: string
- name: speech_tokens
dtype: string
splits:
- name: train
num_bytes: 814869663
num_examples: 258213
download_size: 408821554
dataset_size: 814869663
configs:
- config_name: default
data_files:
- split: train
path: data/train-*