You don't need SF to do that at all. I've done it several times with Vegas, in fact I'd rate it as easier to do in Vegas.

All of that said the outcome can vary from quite good to utterly useless. In smallish rooms the early reflections are well, early and get mixed in with the wanted signal i.e. there's no time or frequency separation between the wanted and unwanted sounds and hence it takes a miracle to sort them out.

You may get some benefit from using simple Eq to tone down the resonant frequency of the room.

Bob.

Yeah, when someone says "room" I get a little skittish, for exactly the reasons you mentioned.

Never tried it in Vegas. Habit, I guess.

Oh, dear. This "echos" (no pun intended) the "use a tripod in the first place and you won't need stabilization" argument.

musicvid, thanks for providing the link!

"Never tried it in Vegas"

Very easy.
Duplicate track, invert phase, add compressor of choice to second track. Adjust compressor paramaters as described. The most critical is the knee however attack and release times also have an effect as does output level. Do not be afraid to play around. One could also use automation to compensate for changes in voice and environment. I've found this trick to work quite well in a large long space made of glass and marble.
What you end up with is a downward expander and it can have other uses.

Bob.

"I've found this trick to work quite well in a large long space made of glass and marble."

That makes perfect sense as the reflections would be more accurate, yes?

It is the delay that is more manageable, thus some forgiveness on the attack time. Works well for concerts in a traditional Catholic sanctuary, (a "large long space made of glass and marble") or in a cave, as I found out second hand from some intel types who picked up the technique on another forum where I originally posted it umpteen years ago.

In addition to the parameters Bob mentioned, getting the threshold just right is key in preserving as much of the dry audio as possible.