metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
splits:
- name: train
num_bytes: 3182575083.5629635
num_examples: 3453
- name: test
num_bytes: 643515185.1730368
num_examples: 864
download_size: 3313267272
dataset_size: 3826090268.736
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*