metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: sentence
dtype: string
splits:
- name: train
num_bytes: 3190083102.449
num_examples: 8473
- name: test
num_bytes: 86360118.544
num_examples: 1896
download_size: 3361103927
dataset_size: 3276443220.993
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*