metadata
dataset_info:
features:
- name: speaker
dtype: string
- name: text
dtype: string
- name: audio
dtype: audio
- name: tag
dtype: string
splits:
- name: train
num_bytes: 3295046908
num_examples: 5679
download_size: 3265748378
dataset_size: 3295046908
configs:
- config_name: default
data_files:
- split: train
path: data/train-*