Is this based on experience or theory?
I'm guessing theory, there are manufacturers who are implementing opto isolating solutions that are far less than a meter, built in specific components, not
just for audio but for enterprise solutions as well.
I implemented fiber optic for my two networked audio systems with great success with 1 meter runs of fiber optic cable for each system with the help
of a good friend who works for a major telecom, and whose very expertise is fiber optic.
The so called "noise" is a very, very easy fix. Too many solutions to note. The easiest and cheapest is iFi
s iPower 9V power supplies, which are quieter than batteries.
There are lots of reasons to use optoisolation in audio gear, but lowering noise isn't one. But your point is valid, I haven't explored the noise performance of new transceivers: my experience is old. I used to build amps and preamps, but also looked at the design of DACs and their challenges, and I've made elementary fiber optic circuits and connections. Optoisolators specifically are really useful for safety circuits. The main audio problem of fiber optic circuits *used* to be the noise created by the receiver, typically a photodiode. Old ones created shot noise about as bad as zener diode. Based on your comment I tried to find comparisons of specific detectivity (D) in current receivers vs those I used years ago to get a sense of noise in current devices. They are much better: a vast amount of research has poured into photonic devices for cameras and telecoms.
Note, the Ifi is an external circuit, so has no application here. You need to isolate the power for the photodiode. That can be as simple as a bypass cap or as complex as a wholly separate supply, albeit tied to the same ground.
Just curious, why did you implement fiber optic instead of electrical connections in your system?