Approach with scepticism!
Well, why not? Shouldn't we always do that with these things?
As to this particular thing, I'll honestly say that I have no idea if any of their claimed benefits are
actual benefits. This is exactly the sort of thing where measurements should help. If it does something , then why aren't they
showing that it does something?
Anyway, I have this thought... The designers and makers of USB devices design and make them to work with PC USB ports
as they are. And that is
including the things that may not be perfect about either the specification or its implementation. If not, then surely their USB device is likely to be poor and not worth buying?
But, OK, let's say yes to that, but go on to wonder, "Is it not possible that this device might not make things
even better?" Ahhh... in creeps FUD, undermining our feelings of satisfaction and security in the setup that, yesterday, we thought was just wonderful.
Thank goodness I don't use USB sound, or I might very well be feeling that myself, right now! Certainly there was a time when I would have leapt at this sort of thing

.
$300 would buy a decent soundcard. If I was choosing how to spend $300 on my PC audio right now, that would be my personal choice.
I am a little less impressed with the company when I find that they are selling things like noise filters for PC fans, and even SATA disks (even the solid state ones!). This is where I fall back on my regular position:
Will they make my spreadsheets more accurate? No? Then they won't make my audio more accurate either, thank you very much.
But hey, say I have an internal sound card. What about the analogue circuitry on that sound card and the electrical noise? Back to the earlier assertion: if it is a half decent device, then it is working in an environment
that it was designed to work in, warts and all. If not, then it wasn't worth buying.
Still, though, there is the FUD, and many people will reach for their credit cards. Especially as some these things are not, individually, very expensive.