metadata
dataset_info:
features:
- name: audio_token
sequence:
sequence:
sequence:
sequence: int64
splits:
- name: train
num_bytes: 24175612268
num_examples: 2496717
download_size: 3790160325
dataset_size: 24175612268
configs:
- config_name: default
data_files:
- split: train
path: data/train-*