metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: transcription
dtype: string
splits:
- name: train
num_bytes: 4537829218.874906
num_examples: 10963
- name: test
num_bytes: 266161148.69009447
num_examples: 576
download_size: 4961006232
dataset_size: 4803990367.565001
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*