metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: source
dtype: string
splits:
- name: train
num_bytes: 4288575893.339
num_examples: 5393
- name: validation
num_bytes: 796778288.875
num_examples: 1105
download_size: 5073133367
dataset_size: 5085354182.214001
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*