Source data taken from project page, then split into portions and transcribed.
Ready to use as a validation or benchmark set for speech recognition, speech processing.
Author: Pawel Cyrta
Data taken from [50 Languages project](https://www.50languages.com/language-mp3.php)