When you listen to any professional broadcast, you hear audio that sounds clear and natural and has, mostly, the same loudness. This type of audio has narration that cuts through music, sound effects and noise. It just sounds impressive. But while tweaking the audio levels on your latest video project, you just don't seem to be able to get that same great sound. What's the secret?
Read the full article The Compression Connection.
Start Your FREE Trial Plus Membership To View This Video
Why Become a Plus Member?
As a Plus Member, you'll enjoy:
- Exclusive access to 1,000s of articles, tips, and videos
- Unlimited access to Videomaker Tips & Tricks video series
- Special contests and monthly drawings
- Members only eLetters
- Early online access to the current issue of Videomaker Magazine
- Members only discounts on Videomaker merchandise and more
- Priority status at Videomaker events
- The Expert Hotline: direct email access to our editors. Get answers to questions about any video subject
All for just $24.99 a year!
Welcome to this tutorial on audio compressors for Videomaker, I'm Brian Peterson. I'm going to show you how to use an audio compressor to take a pretty average sounding wav file and turn it into audio that just demands your viewers' attention. I'm sure all of you have watched and listened to television where even a whisper sounds like a roar, and this is particularly true of all commercials. Well, the secret is in using an audio compressor.
Now, there are two types of compressors, there's hardware and software. The hardware kind are great for voiceovers, narration, and they're really not that expensive any longer. So if you do much narration of your own, I'd strongly suggest that you invest in one of these. But for most of us, we're going to be dealing with the filters and plug-ins that came with our audio editing package, so that's what we're gonna be dealing with today. But remember that all the settings that we're going to be talking about apply to hardware as well. So stick on your headphones, grab that mouse, let's dig in.
Here we are in Adobe Edition 3.0 with our file loaded already. Now, this is a 30 second spot for a fictitious appliance store and we've got some noise at the beginning here. Now, this is something I encourage everybody who is recording or attempting to record very good audio, make sure you get a sample of just nothing, the room tone, the outside, wherever you're recording, so you get some really clean nothing. And of course, nothing is always something, and in this case, I recorded this in a room that has a couple computers in it, one has nine fans, so this is what nine fans running looks like.
The reason I'm taking any time talking about noise is because if we were to apply compression to this and the noise, the noise could be increased disproportionate to the signal that we want to keep. So let's get rid of noise first. I'm gonna go up to Effects, Restoration, Noise Reduction. The first thing we want to do is capture the profile, and that profile is going to be capturing just the section right there. Now, it won't change too much, in fact it didn't change at all because I had done that in the past and it's a sticky menu here so it does remember what was last there.
But now I want to select the entire file and this setting right here, the noise reduction level, as you can see, raises and lowers the amount that it's going to impact the original signal. Now, oftentimes, the default setting, which is usually fairly high, works fine, but you need to listen to it, and if it sounds just funky, you know, it sounds effected, just start pulling this back. And so let's go ahead and take a listen. Now, we've got of course that quiet section in the beginning, so we should hear it about now. [Recording playing.]
All right, so you really can't hear much negative effect having it up to 100 percent, so that's a fairly clean signal, or at least noise signal, and it's gonna be just fine at 100 percent. Let's go ahead and apply and we'll see this go just about nothing. There we go, we saw a little effect over the entire wav file but not too bad. All right, so we've got a clean signal now.
What we're gonna try to do is bring up everything so that it still has a natural sound to it. We don't want to squash everything, in fact, this is what a squashed file looks like. You can see I over-compressed and expanded the file. That's the same file, believe it or not, it doesn't look like it, but everything's just maxed out right at 0. Let's take a listen to how hard that drives. [Recording playing.] All right, so it sounds like a used car spot, right. Yeah, okay, so we don't want to go there.
So let's take the process step by step. So when I speak of compression, I'm really talking about three things that are oftentimes confused. Now, most of us are video producers and I think it's quite fine that we think of them as the same thing, but we're gonna be talking about three things, compressing signals which are hot, which we have these higher spikes. They're not really hot yet but they are louder. We're gonna be compressing those down. We're going to be expanding the areas in here which are quieter, and then we're gonna finally be limiting the whole file so that it really doesn't go much beyond about a negative .5 DB so we have a little bit of headroom. So there we go, there is the disclaimer, let's get into some of the specifics of the settings.
Lets' go up to Effects, Amplitude and Compression, and then let's go to Dynamic Processing. Now, I know that it doesn't sound right off like a compressor, especially when down here you see multi-band compressor. We'll get to that, but that's a little bit more advanced, in just a moment. The dynamics processing gives us an ability of looking at the graphic representation. Now, I'm gonna pull this down here, and as you can see, I'm gonna hide this little dotted line right here that goes at a 45 degree angle from left to right. And I'm gonna create just a very flat response curve here. In other words, it's not even a curve at all.
And if we go down here into the summary, we can see that, in fact, it says flat, 1-to-1 above 49 DB, which is this point right here. So what that means is this X axis is the input signal and the Y axis is the amount of gain we're going to provide to the signal. So at this point nothing is happening. To prove that, I'll go ahead and play a quick preview, and we'll hear that it sounds the same when I click on this. [Recording playing.] Off, on. All right, so that proves that nothing's happening yet.
So to apply some very simple compression, and this is a very standard setting, we're just gonna drag this up and give a little bit of a curve so that we've got this arch happening right here, and in some parliaments this is called a hard knee compressor. We could create a soft knee if we just click on spline curve, and see, now we have a softer knee going on right here. It makes the transition a little bit less pronounced, although it's not always the best effect, so I'm gonna take that off for right now. Let's listen to what this does to the signal. As you can hear, it's really pumping up the lower _____. [Recording playing.] All right, you can hear the difference. [Recording playing.]
All right, so this actually is a very usable setting right here, and very commonly, this will be just what you need. So let's go ahead and apply that and see what this – the wav form looks like. So we're gonna click okay. And you know, actually that's not too bad. We've got some peaks coming up above 1 DB, though they are brick-walled or limited at 0 DB, so we're not actually distorting any of the signal. In fact, I'd say we got pretty lucky there.
So let's do a control Z to undo, so we go back to the original setting right here. We want to take another look at a compressor that gives us a little different option. We're gonna go up to Effects, Amplitude and Compression, and now let's go down to that Multi-Band Compressor.
Now what this is going to do, it's going to separate the signal into four separate bands. It's essentially like taking four different compressors and applying them to different frequency ranges, and probably the best place to start, since this is a little bit more complex – no, I take that back, it's a lot more complex, and frankly, you likely can mess up a signal more than you can help it if you really start tweaking beyond some of the presets. So let's start with a preset and see how far we can go from there.
We're gonna click Broadcast, because this seems to do a good job at optimizing all the signals, keeping within reasonable range. Now, let's listen to the signal and see if we're in the ballpark. [Recording playing.] Without. [Recording playing.] All right, so you can hear that there really is a difference. It allows us to hear relatively more bass in there because we're actually reducing a little bit more of the midrange here and here than we are the bass and the high end. So that's why it sounds a little fuller. So if you find that one of these presets works for you, go for it. When you start tweaking around, these are the threshold sliders, you may find that your signal becomes muddier, and when that happens just go back to a preset and see what happens.
Let's go ahead and click okay and see what the signal looks like after this. All right, so it didn't drive it quite as high, we didn't have the gain output set to anything different. But it did bring up some of those lower areas a little bit better. So I think this is a good wav form to do our next step on. So we've got something that's very close.
Now, the next step in overall compression is going to the same submenu here, and we're gonna go down to Hard Limiting, because we want to bring this up and let's keep it max amplitude to just under a 0 DB. We've got minus .5, and boost input – and this is just kind of a guess. We can see that we're about minus 3 here, we're a little bit hotter down here. Let's just do let's say 3 DB and see what that gets us. All right, well, that's not too bad. Now, we're biased a little bit to this side of the wav form, not bad. Let's take a listen and see if it's getting close. [Recording playing.]
All right, that doesn't sound too bad. We were overdriven just a little bit, I had my output on – my main computer out just up but the signal is actually pretty good. So I think we are almost to the point where we would say, let's leave this alone. We've got a good signal. Let's not mess with it too much more.
Now, the only other thing you could do is apply a little bit of EQ and that just starts getting into the sweetening process. But at least you've used a compressor, you've brought up the levels, and the one last piece of work that you might have to look at is a byproduct of bringing up lower levels. Now, I'm gonna zoom into this section right here, which I think is a breath. Ah, yes, there is a breath right there. Let's listen to that. [Recording playing.]
All right, not too bad, but what can happen if you have somebody who is a little bit breathy or just, you know, tends to emphasize their in-breaths. This one's a little bit louder, let's take a listen –. The compressor can brings those up to be an obnoxious level. So what I like doing is highlighting that section, coming back in and then just manually reducing those amounts with another envelope.
So we'll go down to an envelope, and I've got something that I created called a breath cut, and you can just make this out of any one that you want, take a – well, let's get this ______ right here and I'll just make one real quick. We just take these nodes, slide me up to the top, bring this node down here and then click on, to make another node. So we just have like this little half-pipe going on here, for those of you that used to skateboard. And let's go ahead and apply that. All right, so you can see what happened there. Let's control Z that, so that's what it was and I've got this on a hot key for F5, and we'll click okay. And so that brings it down to at least a sub-obnoxious level. [Recording playing.] But it's still there, and unlike the days where digital editing had first come out and there was nobody every taking a breath, you know, I like leaving the breaths in there, they actually do sound believable, but this is at least not obnoxious.
So that's maybe the last step that you would want to take in optimizing your signal. So here you go, you've got an optimized file that's about as loud as it can get while still sounding natural, even the quiet sections in here sound pretty good. So sound effects, music, all of it, this will not get lots in the mix.
That's it for this tutorial on audio compressors. Now, take this new knowledge and go make your audio sound as good as it possibly can. For Videomaker, I'm Brian Peterson.
[End of Audio]