metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: text
dtype: string
splits:
- name: train
num_bytes: 2084570534.588
num_examples: 7647
- name: validation
num_bytes: 478595569.856
num_examples: 1916
download_size: 2400347680
dataset_size: 2563166104.4440002
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*