There are basically two mindsets to audio levels:
- Live audio needs to keep some "headroom", because you don't know what's coming, and you don't want any of it to clip.
- Recorded, produced, and distributed audio knows exactly what's coming and how it behaves, so it intentionally uses the entire scale so that the inherent noise of its (cheap) medium isn't an issue.
Live broadcast (TV, radio, streaming, etc.) is both simultaneously. The mic is live, so you need to treat it with that mindset, but the audience is used to the recorded, produced, and distributed level.
In the digital world, full-scale is full-scale. The maximum possible value is always 1.0 (or 0.9999...). In technical terms, all the bits are fractional, and adding more bits only gives you better resolution in the same total range. If you put a few things together, the conclusion is that your audience expects full-scale.
But the mic is live! You don't know what's coming! How do you make it *exactly* full-scale without ever going over, if you can't predict it?
That's what a compressor is for, followed by a limiter, which is really just a specialized compressor. Essentially, an automatic volume control, with the goal of making its output volume more uniform, and thus predictable. OBS has both, as Filters.
If you adjust it wrong, it's really obvious that you've got something messing with the audio! But if you know what to listen for and what the controls do, and get it right, it sounds completely normal, with a rock-solid output level that you can put exactly at full-scale and never clip.
Compressors are one of the constant-use tools in the pro audio world, so there are TONS of tutorials in pretty much every format (blogs, YouTubes, etc.), from almost every mindset that you can imagine. Pick some that work for you, and have at it!
