metadata
dataset_info:
features:
- name: transcription
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
splits:
- name: train
num_bytes: 3156450413.048
num_examples: 11708
- name: test
num_bytes: 196182968
num_examples: 824
download_size: 2541889628
dataset_size: 3352633381.048
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*