The controls for Audio Leveling are too basic. Different kinds of audio content need different attack and release times, but the most important thing would be some metering to be able to see how much something is getting compressed or limited.
In recording applications (ProTools, Logic, etc) it's typically called Compression and/or Limiting. Ability to control attack and release times would be great, selectable "Auto Gain", ability to "side-chain" the compression (for example you put that on a music track so it will automatically duck the volume when the voice over on another track starts/ends. The most important thing though is some metering to see how much gain reduction is being applied, and input versus output.. To get fancy & simple at the same time, add presets for "Punchy Voice Over", Soft Voice Over, Pop Music, Slow/Smooth music, Sound Effects". Or better yet, enable external plugins via VST!
or provide VST plugin interface to other tools like pro tools, izotope, etc. Trying to automate this for everyone is going to be a loosing a trade off every time. Some will want a preset with hi mid low others will want a gain, attack, release control. What ever you do, make sure to provide a user selectable default override that sticks between projects. Please.