metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: sheikh
dtype: string
- name: duration
dtype: float64
- name: text
dtype: string
splits:
- name: train
num_bytes: 46292804960.064
num_examples: 180844
- name: validation
num_bytes: 9536779257.819347
num_examples: 43626
download_size: 55763993917
dataset_size: 55829584217.88335
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*