metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
splits:
- name: train
num_bytes: 2741051024
num_examples: 39585
- name: test
num_bytes: 249790491.52
num_examples: 5108
download_size: 2921057376
dataset_size: 2990841515.52
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*