Distorted audio

[Bree430]

Hi all. I have been trying for weeks to figure out why my gameplay recorded in OBS always sounds too low in some parts and distorted in others. I read so many posts and tried so many things and nothings worked. We only have 3 tracks set up for the mic capture, source, and desktop. I changed the settings for the audio monitoring and they all basically gave the same results except 1 which took all the sound away. I took off all the surround sound, took off the noise reduction, literally everything I came across on here I tried and it isn't working. I use Davinci Resolve to edit and in there it shows there are 2 audio channels on the track with the gameplay sound capture as well as 2 for my mic recording but I never have this issue with the mic recordings. I played around with adjusting those settings and nothing worked there either. So far when I record from the Nintendo switch, this does not happen. Only from the PC with Steam games so far. I'm hoping maybe there's some new or updated info at this point because the posts I read were older but I cannot really imagine too much has changed I just feel desperate. I am really new to this and I am learning as I go. I watched a lot of tutorials before jumping in but this issue never came up. I'm not sure if I would be able to locate info needed but if there is anything you want to see that could help you help me, please let me know. Thank you!

 

[AaronD]

Welcome to the paradox of live broadcasting! You have to keep it low enough to not distort, ever, not even with unexpected peaks, but you also have to make it loud enough to overcome the deficiencies in the medium that you're broadcasting on. At first glance, you actually have the wrong overlap, where it's both too soft and already distorted. Hence the paradox.

The solution is compression. Essentially a fast automatic volume control that turns the soft parts up and the loud parts down. OBS has one as a filter. Set it well, and you get the loud, clean broadcast sound that most others have, without actually noticing that it's being processed. Set it wrong, and of course it's obvious!

Mic -> Noise Suppression -> Compressor

• Keep the noise suppression first, so it operates on the raw mic signal, which itself is low enough to not clip, ever.
• Follow that with the compressor, which smooths out the peaks, and is then set for enough gain of its own, to bring it up to a decent broadcast level.

As for how to set the compressor - threshold, ratio, attack, release, etc. - look up a bunch of YouTube videos about how they work, and see which method of explanation "clicks" for you. They all work the same, whether analog or digital, or who makes them, so just take a wide swath across all of them.

You might also end up with a limiter as the absolute last thing. That's essentially a super-aggressive compressor, so that the level NEVER goes above its threshold, for any reason. If you do use it, consider it as more of a "safety net" than a continuously active thing. You might bump against it every once in a while, but the normal compressor still does most of the work.

 

[AaronD]

You don't need any of that for a premade soundtrack, because it's already baked into that soundtrack. Just send it out as-is. Only live stuff like mics and instruments need it explicitly.