metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
splits:
- name: train
num_bytes: 678202991.2
num_examples: 2450
- name: test
num_bytes: 282629132.55
num_examples: 1050
download_size: 910374411
dataset_size: 960832123.75
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*