It might be important for us to parse this in terms of the frequencies of individual vords. These three lines of f75r, combined, read as:
qokedy dy sheety qokedy qokeedy qokechdy lol qokeedy qokeedy qokedy qokedy qokeedy ldy yshedy qokeedy qokeedy olkeedy otey koldy
In terms of each word's frequency rank within Voynich B (using the VT transliteration), here's roughly what we're looking at:
(#8) (#34) (#453) (#8) (#6) (#968) (#96) (#6) (#6) (#8) (#8) (#6) (#175) (#335) (#8) (#8) (#90) (#106) (#844)
Within the Naibbe cipher, Voynichese words would be taken by default to be either unigrams or bigrams, with many of the commonest ~100 words (and virtually all of the top 40) representing different ways of encrypting unigrams, aka standalone alphabet letters. So a Naibbe-like parsing of this text could come out as:
A B (bigram 1) A C (bigram 2) (bigram 3) C C A A C (bigram 4) A A D E (bigram 5)
With qokedy, dy, qokeedy, olkeedy, and otey provisionally treated as distinct alphabet letters and the rest of the word types treated as bigrams, where each word would be parsed into a prefix and suffix that mapped to a plaintext alphabet letter.
There are possibly letter assignments here that could yield grammatically and semantically valid Latin, especially if qokedy—which repeats up to four times in a row within Voynich B, consistent with the additive Roman numeral IIII—were taken to be the letter I.
qokedy dy sheety qokedy qokeedy qokechdy lol qokeedy qokeedy qokedy qokedy qokeedy ldy yshedy qokeedy qokeedy olkeedy otey koldy
In terms of each word's frequency rank within Voynich B (using the VT transliteration), here's roughly what we're looking at:
(#8) (#34) (#453) (#8) (#6) (#968) (#96) (#6) (#6) (#8) (#8) (#6) (#175) (#335) (#8) (#8) (#90) (#106) (#844)
Within the Naibbe cipher, Voynichese words would be taken by default to be either unigrams or bigrams, with many of the commonest ~100 words (and virtually all of the top 40) representing different ways of encrypting unigrams, aka standalone alphabet letters. So a Naibbe-like parsing of this text could come out as:
A B (bigram 1) A C (bigram 2) (bigram 3) C C A A C (bigram 4) A A D E (bigram 5)
With qokedy, dy, qokeedy, olkeedy, and otey provisionally treated as distinct alphabet letters and the rest of the word types treated as bigrams, where each word would be parsed into a prefix and suffix that mapped to a plaintext alphabet letter.
There are possibly letter assignments here that could yield grammatically and semantically valid Latin, especially if qokedy—which repeats up to four times in a row within Voynich B, consistent with the additive Roman numeral IIII—were taken to be the letter I.