When it comes to the naming of prokaryotes, the International Code
of Nomenclature of Bacteria (ICNB) is prepared to hold its authors to
a much higher standard (Rule 57a: "Any name or epithet should be
written in conformity with the spelling of the word from which it is
derived and in strict accordance with the rules of Latin and
latinization.) Not only do new bacterial names come with a complete
etymology (including gender), but also a recommended pronunciation.
That said, the ICNB has such strict requirements for naming new
species that microbiologists do not routinely name new ones the way
many zoologists and quite a few botanists do. To name a species under
the ICNB, you first need to cultivate it under laboratory conditions
-- and for many known taxa that anyone would immediately want to name
as a species, this is plainly not possible, at least not yet --, and
then you need to publish the name in one particular journal because
it's not valid otherwise. In fact, the ICNB has an official candidate
status for names, and even that isn't easy to reach.
And are those "rules of [...] latinization" written down anywhere?
In so many respects, the ICZN is well behind the ICNB.
The historical development has of course gone in the opposite
direction -- the Strickland Code even prescribed one single way to
transcribe Greek.