metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: transcription
dtype: string
splits:
- name: train
num_bytes: 5367257273.704724
num_examples: 79191
- name: test
num_bytes: 579733939.4022752
num_examples: 8800
download_size: 5812185768
dataset_size: 5946991213.106999
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*