A straight vocal boost would change the audio profile of the overall scene pretty significantly, so I'm thinking some tool that has less impact would be desirable.

> I don't really know how often it's going to be necessary. Here I am watching this police officer blast some garbled mess

I don't know how it'd fare in this instance but continuous improvements to the Shazam-esque algorithms used by copyright-flagging bots might make this less and less likely to go undetected.