You could also use a hard limiter, setting the threshold by finding the loudest musical peak, and using the fastest attack and release settings. Once you find the loudest musical peak (say at -10dBFS, for instance), set the threshold at - or just above - that peak: -10dB. Then, you can add that much make up gain after the limiter squashes the peaks from the applause that rise above -10dBFS.
A 'volume envelope,' to my understanding, does that exact same thing, but saves a few processing steps actually dialing in compressor/limiter settings other than the threshold (which would be preset as far as attack/release and ratio - if the 'volume envelope' process compresses at some high compression ratio like 20:1 or 40:1 instead of limiter ratios like 100:1 or infinity:1)