metadata
dataset_info:
features:
- name: text
dtype: string
- name: voice_name
dtype: string
- name: audio_tokens
sequence: int64
splits:
- name: train
num_bytes: 2281135432
num_examples: 529195
- name: validation
num_bytes: 20969024
num_examples: 4795
download_size: 394231698
dataset_size: 2302104456
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*