Better Dialogue Audio: Compression and Normalization

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
in this episode we're going to look at two tools you can use to improve your dialogue audio for filming video namely normalization and compression check this out all right here we are in Adobe Audition we're going to illustrate how to use compressor to improve the sound of your dialogue for film and video now I have a sample clip here let me just play through the first portion of this so you can see where we're starting and then we'll show you how to get to the end now one of the big challenges with recording dialogue is that a lot of times you can there can be a big difference between the loudest parts and the softest parts alright so the first issue we have of course is that the dialogue is just way too quiet and so we really need to bring those levels up so the first thing we're going to do is go ahead and select all for this clip and then we're going to come up here to favorites and choose hard limit are sorry normalized - point 1 dB now you'll just use the normalized function function in whatever editor you're using and these same principles apply you just got to find it in your own editor this is what they looked like before and this is what they look like now so all we really did was all of the loudest parts we just increased the volume of everything in essence and we brought the loudest peaks so that they were at 0 DB now if you're not familiar with how to read a waveform monitor like this 0 DB is the loudest part and that's represented here at the top and at the bottom the quietest parts are going to be here in the middle and then of course we have a scale here so you can see it starts at 0 DB gets quieter then we get minus 4 5 DB - 7 - 9 it the last number it shows in this particular one is minus 27 and then perfect silence is going to be minus infinity so you can see the loudest parts are at the tops in the bottom and the quietest parts are going to be right here in the middle alright so that kind of solves one of the problems for us is it now we've brought the volume up overall so let's go ahead and play that now so you can see what that sounds like now one of the big challenges with recording dialog is that a lot of times you can there can be a big difference between the loudest parts and the softest parts ok so now you can see what that sounds like now what we're going to do is go ahead and apply pressor to bring these quietest parts and the loudest parts a little bit closer together so that it's a little bit easier on the ears of your viewers and it's it makes a dialogue a little more present it brings it forward some so let's go ahead down here to our effects rack now before we apply the compressor I want to put a hard limiter in here and that's just really kind of a safety thing if you don't know what it sounds like when audio clips or distorts what you should do and that's all this is saving us from it's kind of a safety thing for that if a waveform tries to exceed 0 DB and go into the plus DB range that's called clipping and in the digital world at least so once that happens what it sounds like is very unpleasant and it starts to distort it has a kind of a buzzing noise your speakers will be over driven so if you don't know what that sounds like and you want to hear what it sounds like go ahead and record on whatever you have to record whether that's your camera or an audio recorder crank the gain up as high as you can or close thereto and what you'll what you'll do is just record a couple of seconds of dialogue bring it back to your computer turn the volume way down on your computer and play it back to hear what it sounds like and you can understand what I mean when I say it's pretty unpleasant so we're putting this hard limiter here just so in case we get a little too aggressive on the compressor settings it's not going to start clipping it'll actually cut it off before then then we'll come back in here and let's apply a compressor we're going to use this tube modeled compressor which compressor you use is not that important I want to use this one because has pretty simple controls and it's pretty simple to illustrate how it works now pretty much every compressor has a varieties these similar settings here they may be called something different some of them have more some of them have fewer in general they all have something along these same lines so we have threshold we have gain which is sometimes called makeup gain and some compressors don't have it but most do there's a ratio which is the compression ratio we have attack and we have release now I'm not going to go into too much detail on attack and release here that's a little bit more advanced and I don't want to spend a lot of time doing that but we can come back and go into more detail if that's something of interest to you but I'm gonna go ahead and set the back to two milliseconds for my dialogue and the release to 100 milliseconds which is the default setting for release let's come back up here to threshold and talk about what that means now when you're using a compressor you get to tell it which of the peaks you want it to or which of the with the wave Peaks here you want it to actually compress and to affect and to reduce the volume on so what you do is you set a threshold so let's say like if we're looking at this waveform here and I come across here I can see I have a lot of peaks at or above this level right here this is minus nine dB but a lot of my dialogue actually falls below this so you can see a lot of this material in here is actually below the minus nine DB range so why don't I set my threshold at minus nine and what that means is that anything that passes this threshold it's louder will be compressed and anything that's quieter than this or does not does not pass this threshold will not be affected so let's go ahead and set that to minus nine okay good enough now what that means again anything above that is going to be compressed or made quieter anything below that will be untouched so let me just bump this up here real quick and show you what that looks like so we've set our again our threshold of minus nine let's apply it and see what happens and bam yeah you can definitely see that any of the peaks that were above minus nine which is right here have definitely been compressed and reduced in size and what that means is they've been made quieter and anything that was below that has been largely unaffected so we haven't really messed with any of that so the next question you may be asking yourself as well that's all well and good but now we just have a quieter signal overall well that's true but there are some other settings we can use to address that now let's go ahead and undo that real quick and let's go back into our compressor now let me explain one of these other settings here okay the next most important setting is this ratio and this is telling the compressor how much to reduce those Peaks that do exceed the threshold so for example I've set it here for 2:1 so what that means in practice terms let's say I had a peak that exceeded - my nine threshold by four decibels so what that would mean is eight seven six five so it would be at minus five it peaked at minus five and that could be this peak right here so this peak right here is at minus five and because it's four decibels above the threshold which is nine when the compressor process is that what it's going to do is is going to take those four extra decibels and squash them down to one decibel so that's it that gives you an illustration of how it's deciding how much to compress so anything that's four decibels will become one decibel and typically with dialogue I like to operate with that somewhere in the two to four range I don't like to get too much more aggressive than that because it starts to sound really unnatural but again here we're even kind of pushing it just sort of illustrate how a compressor works but let's go ahead and run with that setting and again just to show you what it looks like just like before boom it drops all those Peaks that are above minus nine and brings them down a lot closer to the rest of our material now again that question I asked before well that's great but now everything is quieter so how do we fix that so let's undo and let me show you one more setting in there back to our compressor all right we'll bring that over here now we also have this make up gain or in this case it's just called gain so what I'm going to do is run the compressor and fine tune this setting and what this game does or make up gain does is it takes the volume of everything and increases it by the amount we tell it so now that we've kind of squashed those Peaks down we want to bring everything back up closer to zero DB so that it's nice and clear and loud for our viewers so what I'm going to do is run through this clip here and while I'm doing that I'm going to tweak that make up gain setting and what I'm going to be doing is aiming for our Peaks to fall in this - three - maybe - one range I want to get them right in there and of course they'll still be some range so it'll be bouncing around some but that's what I'm aiming for so let's go ahead and give that a go now one of the big challenges with recording dialog is that a lot of times you can there can be a big difference between the loudest parts and the softest parts and oftentimes what that amount - is that the start of the sentence is a whole lot louder than the end of the sentence so one of the things that you can do okay that worked out pretty nicely so we ended up at 5.4 decibels of game so make up game we added another five and a half decibels back after we had compressed everything let's go ahead and apply that so you can see what it looks like now again notice how there's there's quite a bit of range here you know we've got our quietest parts that are down in the minus fifteen range and our loudest parts just barely this one maybe this one come up close to zero but when we apply it okay a lot of things come up a little bit closer so you can see we have a lot more writing up near the the top of the range here and even these quieter things it used to be at minus fifteen or now minus seven so and maybe minus nine so we still have some dynamic range there but we've definitely brought it a lot closer to the top here and let's see what that sounds like now one of the big challenges with recording dialog is that a lot of times you can there can be a big difference between the loudest parts and the softest parts okay and just for comparison let's go back to the original so you can hear what that sounds like now obviously this is really really quiet because we haven't even normalized it but let me go ahead and normalize it first and then let's compare the two so this is the original just normalized now one of the big challenges with recording dialog is that a lot of times you can okay so notice that near the end of that sentence is sort of started to really get a lot quieter here's the compressed version now one of the big challenges with recording dialog is that a lot of times you can now a couple of things I am no master at compression so still kind of learning here if you found something there that you know that could improve that process please let us know if you have any questions go ahead and leave those for us down in the comments below and we'll go ahead and do our best to answer those questions or maybe someone in our little community here will have an answer if you haven't already subscribed go ahead and do so and we'll get you more great videos on how to improve your lighting and your sound for video talk to you soon
Info
Channel: Curtis Judd
Views: 404,328
Rating: undefined out of 5
Keywords: video, sound, audio, sound for video, sound for film, audio for video, sweeten, fix, compress, compressor, compression, normalize, normalization, edit, post process, dialogue, speech, improve, howto, how to
Id: 9kal7soRvT0
Channel Id: undefined
Length: 11min 14sec (674 seconds)
Published: Wed Sep 18 2013
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.