metadata
dataset_info:
features:
- name: transcription
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
splits:
- name: train
num_bytes: 355512712.655
num_examples: 4399
- name: test
num_bytes: 48732115
num_examples: 603
download_size: 403594700
dataset_size: 404244827.655
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*