Having been trained in studio recording, I usually do my own mix-downs for animation tracks, using VS's internal audio tracks as guides. But it occurred to me that some stuff I'm doing for myself using pro software could be done automatically inside VS.
Music and speech (voice-over) are two different tracks in VS. Instead of summing up the signal of music and speech - the app could automatically make speech louder when it exists, and favor music when the speaker pauses.
I'm talking about the effect that's often used in radio stations and often refereed to as "ducking".
To keep things simple, the app could offer three or four ducking algos, from soft to hard.
I didn't check but if overall compression and loudness optimization isn't already used, it should be.