I have these mp3s that I would love to listen to but I can;t get past the plosive over driven audio.
I do have RX5 now so I dropped it into the stand alone, and did the de-hum and denoise thing, but not much help. I haven't watched the tutorial video yet so I'm just banging around. I pull up the spec analyzer and got rid of the nasties, but I just slid the 2 sliders til it was more acceptable. 


I decided since I'm clueless, I would upload the intro to soundclound and I'll see what your approaches would be. I know what I did is not the way it should be accomplished. So, here tis....and thanks
 
https://soundcloud.com/charlyg/intro-to-relaxation

That is a truly dreadful recording. An object lesson in how not to present the spoken word.

It sounds already compressed beyond sensible amounts and like it was recorded without a popshield and the microphone far to close to the mouth. What might help is a de-esser to reduce the sibilance and exagerated "T", "P" etc. plosives plus a hand-drawn automated volume envelope to reduce the huge gasps for breath to silence or near it. A carefully set up noise gate might manage the same thing but as there's little volume difference between the speech and breath I doubt it.

I do have Nectar 2 Breath Control, which will take out the breathing if I figure out how. I got it separated but need to watch a vid to finish..... I can't figure out which vst I have deals with plosives. I have the Izotope Suite but I don't see it. I also looked in all my vsts and I can't find what I'm looking for. I must not know the tool name. I did d-ess and it helped....I also took out the noise. Rx5 has a learn tool that did the trick. I will be watching vids on that as well!

The problem is that the recording is an mp3 it's already had significant material deleted and other compressed so there's not much left to work with.

When doing restoration the single most important technique is to keep in mind that your goal should be to improve the sound, not necessarily make it perfect. There's only some recordings where you can fix everything. Considering the material you have to work with this isn't one of them. Sorry to be so blunt.

Dealing with plosives is pretty much the same as dessing. A frequency-dependent compression process, but usually over a wider range of frequencies.

One way to do it with the tools in Sonar is to duplicate the vocal track. Then either eq boost the frequency range where the sibilance or plosives are heard or manually go through the audio track removing everything but the plosives.

Put a side-chainable compressor on the original track with the side-chain activated by the "plosives" track and adjust the compressor settings until you (hopefully) get a usable result.

It does help if the original audio hasn't already been noticably compressed though, which is one reason for not destructively compressing while recording - once the dynamics have been flattened it can be very tricky or even impossible to undo later.

At least it's just so I can listen, it doesn't have to make the "artist" happy. I thought it might be good practice. The quality can't get much worse!

When I saw this thread the one thing that came to mind was how many mp3s are you talking about? The reason I ask this is it seems counter-productive to intensely edit something to "good enough" status. Does RX5 have a "batch conversion" option in it?
 
In a situation like this (especially if a lot of files), I would be more apt to use a program which has batch conversion capability. It is highly likely those mp3s all have the same issues and were recorded in the same environment. I do not have RX5, but have Adobe Audition and believe some parts are similar. I would use internal settings (per above) to adjust one mp3 to "good enough" and save the settings used by function, then run a batch conversion on the whole set to perform the same steps (is a macro essentially) and save them to another directory. In Audition, even a "background noise print" can be captured from one and used on them all; I "assume" RX5 would be the same.
 
Alternatively, you can drag/drop them all end-to-end in SONAR and work on that track, then export them using the new clip export, but I have not tried this and not sure if bouncing to tracks (needed to embed the FX you do) will keep the clips separated (I think it does, but never tried this). The other reason I prefer a batch program, is Step 1 in any such case is "Normalize to -3dB" so the FX chain used is apples-to-apples (or "good enough") for each file modified.

yes it does(RX5)  and there  are 14 20 minute sessions