metadata
dataset_info:
features:
- name: text
dtype: string
- name: speaker_name
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 22050
splits:
- name: train
num_bytes: 3925190497
num_examples: 17692
- name: validation
num_bytes: 441177332
num_examples: 1966
download_size: 4199026046
dataset_size: 4366367829
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*