metadata
dataset_info:
features:
- name: sentence
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
splits:
- name: train
num_bytes: 9729456696
num_examples: 30577
- name: test
num_bytes: 2078996802
num_examples: 5656
download_size: 11424438928
dataset_size: 11808453498
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*