metadata
dataset_info:
features:
- name: text
dtype: string
- name: audio
dtype: audio
- name: speaker
dtype: string
- name: latin_text
dtype: string
splits:
- name: train
num_bytes: 71311278
num_examples: 1000
download_size: 70725230
dataset_size: 71311278
configs:
- config_name: default
data_files:
- split: train
path: data/train-*