Audio compression questions
 

Hey there. I recorded a choir concert recently and it sounds fine in the timeline, it sounds fine on the video, but they also wanted me to make them an audio CD. I'm having trouble getting the audio to sound good once it's burned to a CD. The Wavs sound fine, but once they're burned to CD...I dunno, something happens it it sounds like what you used to get when you dubbed a recording of a recording with tapes. It sounds like a 128bit MP3 or something.

I've done this before and it's come out fine, so I'm not quite sure what has changed. I don't have audio normalization checked or anything, so I don't understand what burning the Wavs to an Audio CD would be changing. Should I export some other format?

As part of my normal workflow, I import each Wav into Audition and use the compressor plugin to sweeten the audio a bit. Is it possible that I'm doing something different in that step, and it's compressing twice?

Any help would be appreciated.

Thx!

Re: Audio compression questions
 

So if we understand your scenario, you process the audio for the CD differently than you did for the video, and then you complain that the audio sounds different on the CD?

So what does a CD sound like when you DON'T use your "normal workflow" and "use the compressor plugin to sweeten the audio a bit"?

Re: Audio compression questions
 

I would think the compressor plug-in would compress the audio. I personally wouldn't call that sweetening. In fact for something like a choir, which is probably at least semi-classical in nature, I would probably expect the result to be more like "souring." (OK, call me an audio curmudgeon if you like.)

IMHO classical music might need a bit of intelligently done manual gain riding, if the original dynamic range is very large. Other than that, automatic compression might easily ruin the dynamic range and introduce all sorts of pumping and other unwanted problems. Might be just the ticket for club and trance tracks, though. ;-)

Re: Audio compression questions
 

I'd also test without adding the compressor, that's good advice above.

Do render out at 44.1KHz, then listen to that file. You didn't mention how you're getting to CD-Audio, but at some point it must be converted to a 16/44.1 stereo wav, that's the standard. Take control of the process and do your own render to this format, then review the file. The CD authoring software should then just lay it down on the CD without further conversions.

Re: Audio compression questions
 

Once I got the balance I wanted from my various audio sources, I rendered the entire concert out to a CD standard .wav file. I then pulled it into Audition and clipped off each track (adding fades to either end). Because I used various audio sources (house-mix, my own mics, etc) there were some songs that sounded a little weak on the low-end. I used the compressor to boost it ever so slightly, which sounds great when I bring it back into the timeline for the DVD, but when I burn the same file to the CD, something happens.

I realize that part of my problem (and why I am stuck swapping around audio sources from song to song) is that I am a one-man crew, and really this needs someone to figure out levels during the rehearsal and adjust accordingly to the show. Unfortunately I don't have that luxury.

I will bypass any extra compression and just burn the wavs and see what happens.

Re: Audio compression questions
 

Not exactly. I bring these same .wav files back into the timeline to use for the DVD when I'm done. When I burned the test DVD, the doctored audio sounded just fine. The CD turned out differently for some reason.

Re: Audio compression questions
 

This explanation paints a different picture from the original post. It sounds as if you are now asking, "Why did one set of .wav files sound good when burned to a DVD, and bad when burned to a CD?"

If that's the case, two questions come to mind:

1.) are you using the exact same playback system for the CD and for the DVD?,

2.) what do you mean by "different"?

It's pretty hard to speculate about (2.) since you haven't posted any files for us to evaluate.

Re: Audio compression questions
 

This is why there's a whole profession called "mastering engineers". What they do is take the output from the recording / editing / mixing workflow, and translate to CD.

Part of this is resampling from whatever format you have, into the 44.1 / 16 format. And this introduces noise. Which is why they use a noise shaping dither (to put the required noise where it's almost completely inaudible). But that's just the start.

Most of the tools mastering engineers use are only available in the top end range of the high end DAWs (think Sequoia, Pyramix, SADiE, etc.). Since they are expensive, most of us never see them, and therefore don't know what they do or what they are for.

That said, the obvious way around that is to visit an audio post house. They have the tools, and the personnel, to give you that commercial CD look-'n-feel and more importantly, sound.

Alternatively, if you want to learn how to do it yourself, there's a mastering-for-beginners forum over at GearSlutz. And of course, who knows how many books available. It's a big topic.

Re: Audio compression questions
 

"Weak on the low-end..." Usually that would mean weak in the bass frequencies, is that what you're referring to?

If so, a parametric equalizer (which Audition has) would be typically used to bring up bass or whatever frequency ranges seem low, not a compressor.

(I'm not sure if it goes back to CS6, but Audition CC does have a multiband compressor from Izotope. This allows you to compress different frequency bands differently, and can also be used to advantage in some situations, but probably not this one. It's more useful when a freq range is peaky... tame the volume peaks in that range without turning down the lower volume sounds... or boost the lower volume sounds in that range without blowing out the peaks.)

But this multiband compression would be a second choice (in my book), to parametric EQ, which only boosts the volume of a specific frequency range. Apply Parametric EQ in the effects rack, grab the left-most point in the graph, bring it up, move it left and right. Switch the effect on and off. Tweak as needed.

The workflow you described is still a little confusing. *Do* keep your entire workflow at a 48KHz sample rate. The only part of the project that goes to 44.1KHz should be the export for CD at 16/44.1. Audition's habit of converting everything to 32-bit should also be watched; it's good for the Audition project, but you don't want that on exports for CD.

Once you have your 16/44.1 exported file, listen to it! You can do that right in Audition in the Waveform Editor. If it sounds right there, but sounds different when burned to a CD, there's something off in the CD authoring process or software.

All this comes back to workflow, what you do in what software, and what the files are that you pitch between softwares. This can work.

Re: Audio compression questions
 

Thanks all.

Yeah, the one-man-crew/jack-of-all-trades has its drawbacks. This is a charity gig that doesn't pay enough to hire anyone else, however I was trying to make things better than "good enough" if that makes sense. This is my eighth year doing this show, and I've never had complaints, but I always feel that my own standards should always be higher than the client's.