metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: text
dtype: string
- name: phonemized_text
dtype: string
- name: speaker
dtype: int64
splits:
- name: train
num_bytes: 350311112854.716
num_examples: 653109
download_size: 316422690108
dataset_size: 350311112854.716
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
language:
- de
size_categories:
- 100K<n<1M