umm just provide a loudness level that you are ok with as a reference, and by monitoring the loudness level if sound goes above it, just lower it. this is possi

umm just provide a loudness level that you are ok with as a reference, and by monitoring the loudness level if sound goes above it, just lower it. this is possible to program in linux and straightforward also. no need for machine learning. but yes machine learning would be awesome. ontopic this idea is great. you can implement it via formula for loudness that takes distance and decibels as variables.