metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcripts
dtype: string
splits:
- name: train
num_bytes: 42165747125.1
num_examples: 45000
- name: test
num_bytes: 4670187923.9
num_examples: 5000
download_size: 42902259467
dataset_size: 46835935049
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*