Here’s the short answer: It’s not actually louder, it just sounds that way.
The long answer: There’s two ways to descibe loudness or (as the engineer types properly refer to it as) modulation. There is peak modulation and there is average modulation.
First, picture in your mind a VU meter. You’ve seen those before, on tape recorders and the like. Digital, analog, it doesn’t matter. The thing to remember is when the needle (LED, LCD, etc.) crosses from the black (blue) into the red, that point is called 100% modulation. Anything over that 100% will overmodulate the signal and cause clipping and/or distortion.
Now, peak modulation is what the signal is doing every millisecond. It would be very difficult to read those peaks on a meter, not to mention expensive, as the circuitry of a peak meter is very complicated compared to the other type of meter, which is an average modulation meter. Those are the types we’re most familiar with.
Still with me? We’re almost there. So, if your average meter is timed for 1/10 second, and within that 1/10 second it receives samples of 75% modulation and 125% modulation, the meter will show exactly 100% modulation. Easy enough? Then here’s the answer to your question:
Commercials are designed to have greater peaks than average programming!
A broadcasting outlet cannot modulate their signal greater than 100%. Therefore, they utilize equipment such as the Optimod[sub]TM[/sub] from Orban Electronics. In it’s simplest form, if the signal is too loud, it makes it softer, if it’s too soft, it makes it louder.
Fine, in theory. But notice the difference between a TV show and commercial music. TV music is dynamic. It varies, most of the time soft behind dialogue. Commercial music is loud. It peaks very high. The announcers usually punctuate every word.
The Optimod (or other processor) is there. It tries to maintain order. But the low peaks on the show can only be brought up so far, and then when the high peaks hit, it’s all the poor processor can do to keep it within FCC limits for levels. So it’s “highly compressed.” You’re hearing a signal that maintains at 100% vs. a signal with dynamic range to it. More peaks vs. less peaks, run through a processor. The more peaks version will “sound” louder.
Now if you’re talking a cable channel which have no such limits, all bets are off.
Hope you’re still awake and understood that explanation.