metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 24000
- name: sentence
dtype: string
splits:
- name: train
num_bytes: 2235231169
num_examples: 5000
download_size: 2203027948
dataset_size: 2235231169
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
language:
- de
13 hours of audio data randomly selected from the data of the speaker "Eva K"