Writing a fake Voynich by rules: copies, habits and the link between words

[Mauro]

Could you post a sample of the generated text, possibly divided in sections? Thanks.

There are samples in the repository, but I would be more interested in pseudo-Voynichese generated without mixing Currier A and B or even different "dialects" of the same Currier "language"... I don't understand what is the point: the result can't be compared with any real Voynichese.

I understood each section of the manuscript with its own set of parameters, without mixing languages?

Edit: I found the samples, yes they are generated by learning on the whole manuscript. Separate sections would be best.