metadata
dataset_info:
features:
- name: audio_token
sequence:
sequence:
sequence: int64
- name: text
dtype: string
- name: speaker
dtype: string
- name: book
dtype: string
- name: audio_duration
dtype: float64
splits:
- name: train_clean_360
num_bytes: 3310792673
num_examples: 116451
- name: train_other_500
num_bytes: 5367393098
num_examples: 205035
- name: train_clean_100
num_bytes: 931052544
num_examples: 33232
download_size: 1522957345
dataset_size: 9609238315
configs:
- config_name: default
data_files:
- split: train_clean_360
path: data/train_clean_360-*
- split: train_other_500
path: data/train_other_500-*
- split: train_clean_100
path: data/train_clean_100-*