metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: label
dtype: string
splits:
- name: train
num_bytes: 40306259557.384
num_examples: 936146
- name: test
num_bytes: 705562866.39
num_examples: 16197
- name: eval
num_bytes: 694872767.769
num_examples: 16197
download_size: 35968213162
dataset_size: 41706695191.543
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: eval
path: data/eval-*
license: cc0-1.0
task_categories:
- automatic-speech-recognition
- text-to-speech
language:
- de
pretty_name: Common Voice 22 German Split
German split of Common Voice 22. cc0 license