metadata
dataset_info:
features:
- name: text
dtype: string
- name: speaker_id
dtype: int64
- name: audio_codes
list:
list: int64
splits:
- name: train
num_bytes: 306522834
num_examples: 5589
download_size: 48312334
dataset_size: 306522834
configs:
- config_name: default
data_files:
- split: train
path: data/train-*