Certainly this will work best for recordings that used a single microphone for the vocalist. And also certainly it will be most effective where only the vocalist frequencies are reduced. That can be done after the reduction, which will probably allow boosting the non-vocal frequencies. So the operation will be a bit of art as well as science.

 

The PAiA Vocal Zapper seems to be what you asked for. The link goes to a PDF of the full manual, here's a screen capture of just the schematic: