metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: text
dtype: string
splits:
- name: train
num_bytes: 11468757335.587
num_examples: 32901
- name: test
num_bytes: 1111697749.158
num_examples: 3139
download_size: 18101693665
dataset_size: 12580455084.744999
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*