metadata
dataset_info:
features:
- name: file_name
dtype: string
- name: sentence
dtype: string
- name: continuation
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 49238786892.428
num_examples: 26318
- name: valid
num_bytes: 726591416
num_examples: 562
download_size: 47306757996
dataset_size: 49965378308.428
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*