metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: label
dtype:
class_label:
names:
'0': eins
'1': ja
'2': nein
'3': verbinden
splits:
- name: train
num_bytes: 1036620540.45
num_examples: 7981
download_size: 592054581
dataset_size: 1036620540.45
configs:
- config_name: default
data_files:
- split: train
path: data/train-*