Wondering why a simple signal can't be sent to a receiver, especially those with noise max limits built in already, to restrict the volume.....
That's sort of what Dolby Digital does. The normal level of dialogue (an average voice...not shouting, not whispering) is set as a standard. Levels are held to a point that matches that....a standard voice always will come out at the same level, every time, on every channel.
The "magic" is, that standard level can be TRANSMITTED at different levels, to allow for headroom in the system, allowing for more headroom during, say, an explosion or whatever. The DD bitstream carries metadata that pulls the volume back to the correct level right in your receiver.
The second part of the "magic' is, the audio is sent with full dynamic range, but the producer can specify how to restrict the dynamics at your receiver, and the bitstream carries a control voltage to your receiver that pulls up the quiet sounds, and lowers the high volume sounds, all centered on the standard "normal voice level" dialogue. So, when you listen in "Midnight Mode" (or, whatever your receiver maker calls it), the dynamic range is reduced. And, if just listening to stereo or mono from a converter box, it can be restricted even more. The guy with the big HT system still hears full dynamic range, but you can set your own system for limited dynamics (still artistically-controlled by the producers), and it all comes from one transmitted digital audio stream.
They also tell your receiver how to downmix the 5.1 (or, whatever) to stereo or mono, under the direction of the audio producer.
Dolby also has methods that we use to measure "Equivalent Loudness", and that is what really needs to match. A quiet scene, going to a normal commercial, is always going to be jarring, and a commercial leading in to a loud scene is also going to stand out. Both sound "wrong", but are really being sent at the same "volume" level, as far as any measurements (besides "Loudness Equivalence") are concerned.