metadata
dataset_info:
features:
- name: transcription
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
splits:
- name: train
num_bytes: 1762230771.099742
num_examples: 40694
- name: test
num_bytes: 94825239.6342581
num_examples: 2142
download_size: 1847462880
dataset_size: 1857056010.734
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*