13-08-2025, 05:50 AM
I ran your sample texts ( naibbe_Cleaned_52_01_10_word_lines.txt + naibbe_Cleaned_52_02_10_word_lines.txt - 25kb ) and the "original" Voynich text (ZL3a-n.txt) through a language-independent morphological parser. The distribution of the top 25 prefix/suffix pairs is quite simliar. Of course, we are not dealing with a natural language here, but at least the same material is being compared.


![[+]](https://www.voynich.ninja/images/collapse_collapsed.png)