metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: label
dtype:
class_label:
names:
'0': arabic
'1': dutch
'2': french
'3': korean
'4': mandarin
'5': portuguese
'6': russian
'7': spanish
'8': uk
'9': usa
splits:
- name: train
num_bytes: 17639326454
num_examples: 20000
- name: test
num_bytes: 258781838
num_examples: 300
- name: validation
num_bytes: 259776838
num_examples: 300
download_size: 17374935205
dataset_size: 18157885130