Can tesseract recognize sequences of letters in a image that are not necessarily real words nor bound to any human language?

Question

I am trying to do that in Python with tesseract, but it seems to depend on the language to be able to deduce the characters (and that makes sense).

It is a sequence of 14