metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: sentence
dtype: string
- name: language
dtype: string
splits:
- name: train
num_bytes: 5410456276.2560005
num_examples: 67138
- name: validation
num_bytes: 465801573.27
num_examples: 3755
- name: test
num_bytes: 470638482.26
num_examples: 3385
- name: test_lg
num_bytes: 393195163.3179557
num_examples: 2828
- name: test_nyn
num_bytes: 55999446.010044314
num_examples: 557
download_size: 6329160988
dataset_size: 6796090941.114001
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*
- split: test_lg
path: data/test_lg-*
- split: test_nyn
path: data/test_nyn-*