metadata
dataset_info:
features:
- name: path
dtype: string
- name: audio
dtype: audio
- name: transcription
dtype: string
- name: lang_id
dtype: string
splits:
- name: train
num_bytes: 6699328186
num_examples: 36547
download_size: 6218752524
dataset_size: 6699328186
configs:
- config_name: default
data_files:
- split: train
path: data/train-*