metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: speaker_id
dtype: string
- name: formatted
dtype: string
splits:
- name: train
num_bytes: 6061594836.789303
num_examples: 12515
- name: valid
num_bytes: 191900541
num_examples: 540
download_size: 18225043353
dataset_size: 6253495377.789303
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*