metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: transcription
dtype: string
splits:
- name: train
num_bytes: 113083402048.688
num_examples: 826029
download_size: 114088653869
dataset_size: 113083402048.688
configs:
- config_name: default
data_files:
- split: train
path: data/train-*