metadata
dataset_info:
features:
- name: Chemin
dtype:
audio:
sampling_rate: 16000
- name: Langue
dtype:
class_label:
names:
'0': '0'
'1': '1'
'2': '2'
'3': '3'
'4': '4'
- name: input_values
sequence: float32
- name: attention_mask
sequence: int32
splits:
- name: train
num_bytes: 18429503974.38
num_examples: 17574
- name: test
num_bytes: 4634135281.78
num_examples: 4394
download_size: 13665667046
dataset_size: 23063639256.16
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*