Like so many things audio, my best guess is that anybody who tries to definitively answer this question "yes" or "no" really has no idea.
I can say from personal experience that speaker break-in does exist. But I can’t define with any credibilty the scope of products or technologies that it applies to.
In my case, my Audeze LCD-3 and LCD-XC headphones both sounded horrible when I first got them. In particular, the Beatles "Come Together" had such bloated bass that I couldn’t stand to listen to it. In both cases, cycling white noise and looped music through them for a week fixed the problem. After the break-in, both had that exquisite Audeze low end, fast, detailed, and extending effortlessly down into the 20Hz region. Far too dramatic a change to be dismissed as "ears acclimating" (really!) or (if I hear this BS one more time, I’ll scream) "confirmation bias."
Then, a few years later, I came into possession of that same pair of LCD-3s, which had not been used in over 6 months by their current owner. The fat bass was back and when compared to my LCD-XCs, was exactly the way I’d remembered it. I literally could not listen to them playing most music.
But after another week of break-in and the two units sounded pretty similar again in the lower registers.
Anybody who kneejerks about this happening because I ’wanted’ it to occur is merely displaying their own confirmation bias. The difference was repeatable, and was so profound that it would be impossible to imagine -- even if I hadn’t A-B’d the two units before & after the second LCD-3 break-in. To put things into perspective, the difference was on the order of turning an old-time receiver’s bass or treble tone control knob from 1 to 10. Real, significant, and repeatable.
One last note: After owning the XCs for a while, I had the drivers updated by the factory to a newer design. When they came back -- you guessed it -- big, fat, painfully bloated low end. Another break-in and problem solved.
To sum it up: Audeze planar magnetics, at least the models I owned, undoubtedly require break-in in order to be listenable. Furthermore, leaving them unused (at least the LCD-3) for a long time created the need for another break-in period.
I have no idea why this was the case, although there are plenty of credible explanations. But it was the case. It wasn't in my head.
I caution against trying to extrapolate my experience to other types of components, to other planar transducers, or even to other Audeze planar-magnetic phones. All I can say is that audiophile break-in, in my experience, is inarguably real in at least some very popular components.