metadata
dataset_info:
features:
- name: wav
dtype: audio
- name: __key__
dtype: string
- name: __url__
dtype: string
- name: transcript
dtype: string
splits:
- name: train
num_bytes: 1483600251.53
num_examples: 1545
- name: test
num_bytes: 378433471
num_examples: 388
download_size: 1850983991
dataset_size: 1862033722.53
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*