I have two analogue mic inputs. One is a noise insensitive mic with a bad frequency response and one is a conventional mic. The application is speech transmission. The noise insensitive mic will be used in situations where the conventional mic receives too much background noise are does not record anything or the speech recorded is significantly lower than then other mic. The noise situations have been covered but not the latter.
What I am looking for is to compare the energy of the two mics and switch/mix them based on the differences. If the conventional mic does not record anything or the energy of the recording is x% lower than the recording of the noise insensitive mic the noise insensitive mic will output the speech.
I tried subtracting the signals and use a tolerance block to output a one when the noise insensitive mic should be enabled, but I did not get a great result. I used envelopes etc. to create stable positive signals for the subtraction but to no avail, I am not really educated on the workings of the dynamic processors.
Does anyone have a suggestion or maybe there is a useful block I've missed.