jason from volo wrote: Fri Jul 16, 2021 8:53 am
uglysound wrote: Fri Jul 16, 2021 7:50 am
I really enjoy making audio collage, but sometimes only have lower resolution material to work with, like 128kbps mp3 quality. Are there any tricks to take some of the gross sheen off of that stuff?
Yikes. Once audio information has been (overly) compressed some of that information is kinda lost forever. You could use a program like Audacity or similar to try to reduce or filter out some of the worst affected frequencies (thinking things like cymbal splashes and the like), but if not done right that could make things even worse.
Is there any way to get the original uncompressed files?
Maybe someone else has better ideas....
< edit > Two more thoughts: depending on the "gross sheen", think of it as a feature and not a bug. Or, layer another audio track over it that masks the offending sheen. < / edit >
There are plugins that could probably do some of this - Izotope RX and Soundtheory Gullfoss come to mind - but: you'll spend a ton of effort to get at best 20% of the result you want.
Jason's edit is better and cheaper advice.
In similar situations, I usually try to find a way to FIUM it (fuck it up more), either to mask or take advantage of the problems in the source.
Example: last week I was working on this thing where I wanted to use a bilingual station ID segment from an Afghan radio station as a background pitched-down spoken word layer, but all I had was a 96kbps MP3. Because it was voice only, I hard band-passed it like it were going through an Auratone Sound Cube (300Hz and 3kHz, roughly?), doubled the track, hard panned the two, and pitched the right channel down by 3 or 4 cents to give it a little bit of distinctiveness, space-wise. Whenever there was a noticeable MP3 artifact, I'd lean into it by timestretching the whole stereo pair a bit.