metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
splits:
- name: train
num_bytes: 2184790832.5766463
num_examples: 2176
- name: test
num_bytes: 426536584.3624535
num_examples: 273
- name: validation
num_bytes: 420520220.9306072
num_examples: 272
download_size: 5721927764
dataset_size: 3031847637.869707
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*