Is it possible to implement this on a more global level ? The LingQ Library has some texts wich has accent marks as well. It would be cool if words with an accent mark would count towards the same word for Ukrainian and russian. Because a text with accent marks will show as 100% unknown words even when you actually know most of them.