Equalising many audio clips

[biscuits9]

Hi. Recently new Audition user so apologies as this is probably basic and I’m not going to use the correct technical language 😀

My current project:

I am putting together a playlist of songs (not whole songs, rather a couple of minutes of each). 

Ive then exported the whole mix as an MP3 and popped it into a premiere pro project, where I’ve added videos etc. 

What I need advice on:

I would like the peaks of each of the individual clips to be roughly the same volume. At the moment, in Audition, I’ve taken each clip individually and set the max peaks to -3db. 

Would that work in theory? Should my exported MP3 mixdown be around the same volume throughout now? 

Now it is in premiere pro, it seems to have imported in and automatically gone to 0db peaks. Should I then use premiere to set max peaks for the MP3 back down again before exporting the final mp4? 

[SteveG_AudioMasters_]

There are a couple of things: firstly, MP3 is not the format to be doing the processing in at all. You've encoded your mix from the wav file that Audition would have started out with (even if you thought you were putting MP3 files in, you weren't; they were decoded to wav before Audition would do anything with them), and then you've re-encoded this to MP3 again, and then when it's got to Premiere it's been decoded a third time, had the levels messed with and then you want it as an MP4? I fully expect it to sound pretty bad after this.

Secondly, whilst this methodology will certainly set the peak levels to be the same, it won't necessarily have the perceived volume the same. The degree of discrepency depends mostly on the dynamic range of the materials; Audition does have a way of levelling out perceived volume, but I don't think it's likely to survive your present processing regime...

What you should do to preserve what quality you can is to export the audio as a wav file. This is uncompressed, and won't degrade - also you could volume-level it first without it getting messed up. You import the wav file into Premiere, and when you've got everything the way you want it, you then export it to MP4. This way you only do one decode of the MP3s you started with, and no more encoding of anything until the last step. The basic principle is always that if you are going to encode anything to a lossy format (MP3 and 4 are both lossy), then you should start out with the highest quality original, to minimise degredation. With MP3 files, especially if they are only 128k ones, three decodes will make them sound sort-of 'plastic' - nothing like the originals at all, and not pleasant to listen to.

As for the loudness, when you've got your individual files open, you can drop them onto the Match Loudness box (if you can't see it, you have to enable it in Audition's Window menu) where I just dropped a single file, select a processing option (try the one in the screen-grab below) and after you've run it, save the modified wav files, and those are what you use in Multitrack, and the basis of your export.

[biscuits9]

Hi Steve

Many thanks for taking the time to reply. I think that makes sense - I will give it a go!

Might pop a reply back if anything else comes up but appreciate the advice.

Now off to learn some more and experiment...

[biscuits9]

Hi again Steve,

I think I am getting there. One last question (having followed your match loudness steps)

I had previously (before chatting to you) imported all my files and created my multitrack. I have now taken those original files (top left) and dropped them into the match loudness pane and run through your steps. You can see the results bottom left.

Are those new matched files automatically subbed into the multitrack? Or do I now have to go through each clip on the multitrack and swap it with the new matched loudness file? 

If so, what is the easiest way to do this? 

This was the only bit of your instruction I didnt quite follow - 

"...and after you've run it, save the modified wav files, and those are what you use in Multitrack, and the basis of your export"

thanks again

[SteveG_AudioMasters_]

If you look at that list of files, they've all got asterisks beside them. That means that they've all been modified, but none of the modofocations have been saved.

Now without trying this I'm not absolutely sure, but if it goes the way things usually do, then your mixdown will have the changed levels, but if you don't save the files when you close the session, you'd have to go through the whole process again to get those levels back. If you just close the session, the files will be at the original levels.

I hope that that is a slightly better explanation of the last bit. Yes, you have ideally to save the files, but from your screen-grab it looks as though the changes are implemented in the open session. But as I implied before, I would have sorted all this out before even starting the session, and saved the results before placing the files. But that's just me - there's no obligation to do it that way around.