Transcription
[Applause] Having gone through the mixing process of the drums for this track, let's take a look at the guitars, bass, and vocals. And as ever, it's worth bearing in mind that your end product is always going to be limited by the quality of your source tone. So remember that you've got to be working with the very best tracks you can in order to get the best results. So a lot of the techniques you'll be seeing me apply here today are fairly minimal, and that's just a reflection of the fact that the tracks already sound pretty decent.
Also, it's worth noting that I do use a kind of top-down mixing approach, as we touched on earlier. So although some of the moves seem very minimal on these tracks, don't forget there is a bit of extra top-end and bottom-end being added on the instruments—not the vocals. The tones we've got to work with here for the bass are a DI channel and then an amp track, which is a kind of distorted bass tone. It's just kind of in keeping with what I normally do for my bass tones, and this was recorded by me with my Diggle NG2 4-string model in E standard. So I didn't actually have to do very much to the bass guitar to get it to sit into the mix really well.
Here, I'm gonna start by disabling all of the processing this time, and then I'll show you what each bit's doing. But, you know, firstly, I think it's worth talking about achieving the best consistency that you can at the playing end. As you can see here from the DI, that's with no compression, you can see I've really aimed to get my playing as consistent dynamically as possible. And this is kind of crucial to me to be able to get a low-end, especially on the bass guitar, that feels really consistent, and you don't have to hyper-compress to try and get it to sound consistent. But I mean, you're never going to be able to make up for inconsistent playing through compression, because just the timbre of every single note's gonna be so different.
You can also see here that for the verse section, I've played a much lower dynamic, and and just like with the drums, it's really crucial, I think, that if you want something to sound fairly soft, to track it that way. So you can see here that the DI drops down significantly in volume, though the distorted amp track doesn't so much because the distortion that is running through. So let's just take a little listen to the DI first—the intro—and then the amp track. [Music] So the amp track is using, actually funnily enough, a stock Logic plug-in; it's called Grit. It's like a, essentially, a recreation of a Pro Co Rat pedal, and I'm running that straight into an impulse response that I made. And essentially, that forms the bulk of my tone on this track. I'm just kind of blending that with the DI to a small extent. But on the DI, I kind of wanted to give the attack character a little bit more juice, so I thought it'd be nice to compress it, and I thought we could use the Smash-and-Grab compressor. I set it to the overhead mode, which gives a relatively transparent character, and I used it in smash mode, which gives it the kind of more controlled transients. So I'll show you what that's doing on the channel here. I'll start with it bypassed and then I'll initialize it. [Music]
So while we've got them at hand, I actually use the beef and air controls a little bit. So I'm boosting up the beef a little bit to give a little bit more sub-low, and then I'm cutting the air just to trim off a bit of the finger noise that you might be getting, or pick noise perhaps—just that kind of very obviously DI-2 part of the signal—just give it a little bit more around the top. So I think this actually worked out really well on bass for this. The other thing I've done is actually I've used a bit of saturation. I found that the saturation in tighten tape mode, in particular, sounds really cool on the DI, so I'll show you what that does. Just as an aside, but I'm not using it to a particular extreme. [Music] So you could hear, is it got really aggressive there—it starts to sound a lot like a, I call it, distortion pedal effect almost. It's well worth bearing in mind for, yeah, for use in the future. You could even try running that into a cabinet impulse response perhaps, just to see what happens.
And so once the compression was done, I still felt like I wanted to trim the peaks down a little bit more. So similarly to what I did with the tom tracks earlier, I'm using a limiter here just to trim off a bit of the flap from the front end. It sounds like this. [Music] It's a really subtle difference; we're talking about two to three dB if that even, but it just makes sure that everything's kept exactly in check. The amp track didn't really need any processing on its own, apart from I just, I thought I could hear quite a kind of strong resonant note here about 120-20 Hertz. I'll show you what that sounds like if I boost it, just to show you what I mean when I say strong resonant note. [Music] It's quite common when you just record a bass guitar, or any guitar for that matter, that you start to get these certain frequencies that start to dominate, and you can hear them ringing out on top of everything else. So when you can identify those, it's often worth cutting a little bit of it out of the signal just to maintain a more clean frequency response.
I find that sometimes you might want to reduce the mid-range in general on an instrument and then realize that actually it's just one really narrow band of frequencies that are peaking, and using a kind of traditional, fairly wide Q to cut that area would be reducing a lot of the good information at the same time. So I try and keep the Q's quite a small amount when doing this kind of work, in order that you're not really affecting the general EQ curve of the instrument. Looking at the actual bass bus, you'll be able to see the processing which I've done here, which is being done by two different FabFilter plugins. And I'm high and low-passing, as one typically does on most channels, cutting a little bit of the strong resonances that I was getting around about 150 Hertz, 165 Hertz in this case. There's a very strong note in the bass there, especially with it being tuned E, that's quite high up in the spectrum. And instead of having the kind of almost into the upper bass region, like I should say, being too prominent, I thought it better to cut there and instead fill it with a lot more sub. And you'll see over here on this multiband compressor that is only instantiated on the low end, I'm doing a similar thing—actually, in fact, it's almost exactly the same—only being done with a shelf boost here for the low end.
So essentially what the Pro MB is doing is it's compressing—it's drastically reducing the dynamic range of the low frequencies in two different bands—and as far as the kind of 150 Hertz regions goes, it's cutting them, and below that it's actually boosting it to give us a really nice solid sub. So I'll start by showing you what the EQ does, and then I'll bring in the multiband on. [Music] And the keen-eyed of you would have noticed that I'm getting a kind of similar smiley face kind of thing happening here on the frequency chart, like I was getting with the toms, the kick, the snare. There's definitely something in that—it's something which I look for in a lot of instruments. And then bringing in the multiband compression sounds like this. [Music] There's just a sense of, again, tightened up when you engage the multiband compressor, because you know it's not just a fixed EQ curve. So it just sounds like there's a lot more consistency between the notes. There's no EQ that, there's not frequencies being cut when it's not necessary. Instead, like all the notes retain the same amount of body, and the sub-low sounds very kind of pinned and non-dynamic, which I think is good. You want your bass guitar to have a very consistent low-end underneath the rest of the instruments. Now I found that with that I was pretty much done as far as the bass processing on this track. The lower you tune a bass, I feel like the more problems you run into with managing the low-end, but in E standard it's, er, it's really quite easygoing. So as a relatively easy bass track to work with.
Moving on to guitars, there are essentially three different pairs of guitars. We've got a pair of main rhythm guitars which are just playing power chords. We've got some overdubbed guitars which are playing a kind of arpeggiated and wrung-out pattern over the intro and chorus. And then we have a pair of non-symmetrical guitars which, which kind of play off each other during the verse. Let's start by looking at the rhythm guitars. I tracked these with a PRS Custom 24 with Bare Knuckle pickups, and again, being in E standard, you've got a lot of options open to you in terms of tone. So first of all, let's take a look at the rhythm guitars. These tracks here were generated using an amp sim which is made by Scuffle Amps. I think it's a really great amp, so especially for higher-tuned stuff where it's not really all about super high-gain tones. I forget actually which model I use, but I printed it down; I ran it through some custom-made impulse responses that, that I made actually, funnily enough, here at Middle Farm, and I think the tone came out really cool. And this is using a PRS Custom 24. However, the tone was really quite dark. So you can see here the EQ, which is the only processing I've done to these guitars, is reducing a lot of the kind of mid-range emphasis and bringing in some extra top-end and a little bit of extra low-end too. So I'll show you what the guitars sound like with no EQ and then what they sound like with a bit of EQ applied. [Music]
You have to resist the temptation to feel that, that it sounds too scooped, whether you do that because whenever you compare back against the kind of darker and more mid-range source, it can be really difficult sometimes to appreciate the new brighter version. It can often sound a little bit harsh by comparison, but in this case, that mid-range spike was causing a lot of issues once you chucked it into the mix on top of everything else. And applying that EQ allows that to sit amongst the other instruments, especially the drums, which do occupy like way more top and bottom-end information. It's really important when you're EQing to not lose sight of the fact that there's probably going to be a draw to the darker, more mid-range version. Often times you'll find that by the time you've tweaked something to sound really good in a mix, it might sound a little bit more harsh than the un-EQed version. However, if you don't directly A/B it, you don't think that way. It's just one of those things, like I say; ears naturally tend towards the kind of smoother version. So with that EQ carried out on the guitars, I found there wasn't really any need for any more processing. I mean, the guitars are, they're really not very dynamic; there's no palm-muting which is causing excess low-end flub around, and I made sure that as much as possible the source tone was really, you know, well-played and well-captured, so that, so that wasn't really anything to fix in the mix.
So to move on to the overdubbed guitars, here we've got a pair; we've, they're playing the same thing, but I did one side tracked with a Strat and the other with the same PRS that I tracked the rhythm guitars with. So to get these to sit on top, I find that in general when trying to sit overdubbed guitars on top of rhythm guitars, you have to extract quite a lot of low-end; so you have to remove a lot of low-end for it not to kind of just add way too much muddy frequency wash in the mix. So you can see I've got a very drastic high-pass filter going on here; it's like 200 Hertz, and then using just purely subtractive EQ to cut a few of the more harsh frequencies out. Something I often find, especially with guitars, is if you high-pass them quite aggressively, it kind of uncovers this need to cut all sorts of frequencies up in their kind of upper mid-range and treble, just because otherwise it gets really kind of thin-sounding. So you can see here I've used subtractive EQ to cut all sorts of junk out of the guitars up here. I'll show you what the overdub guitar sounded like on their own, and then I'll bring in the EQ to show you. These were using a different cabinet impulse than the rhythm guitar, so they needed a little bit of extra work just to play nice alongside the rhythm guitars. [Music] [Applause] [Music] [Applause] [Music] So you can hear I've carved away a gigantic amount of those guitars, and it sounds extreme on its own, but when you put them alongside the rhythm guitars you'll hear why, because there's just no need for them to have any body. All you really want to hear is like the chime and resonance of the notes ringing out, and it's not even about being super distinct, but just giving this huge wash of sound when combined with the rhythm guitar. So I'll show you what that sounds like with the two channels together. [Music]
So that's what I've done in terms of EQ on these overdubbed guitars. I also thought it'd be nice to add a little bit of echo to them, and I've got a kind of ping-pong going on here with a quarter note on one side and an eighth note on the other, just to give an extra bit of movement to this guitar and make it a bit washier. I've used the studio tape mode in this plug-in called Echo Boy, but notice I've also filtered out quite a lot of the low and high end to make it not become too indistinct. Too much high-end in delays, especially on guitars and vocals, I can find, I often find it just a bit distracting. You get too much repetition of the attack frequencies, and it just doesn't sound very good to my ears. So plenty of high and low cut. So with the echo on there, [Music] it's just a bit of extra texture there. And then finally, I noticed that there was a couple of, well, it was a frequency in particular—this one here, 3.3K—that was a bit harsh. So I'll show what it sounds like without, then I'll engage it, and hopefully you'll hear that that just again tames some of the bite on the guitars—it's kind of a whistling frequency that I'm just not very keen on. [Music] This little notch—most applied across basically everything in the mix for me—adds up to tracks that are way easier to make louder in the mix, because those kinds of harsh whistles and resonant frequencies just eat up so much space once you make the track that contains them loud in the mix. So for me, those are some of the most important moves in my mixing actually—just finding the right frequencies to cut in these very narrow notches, in order that as a whole, that track then becomes something which is way less aggressive on the ear, and therefore you can make it louder and more present in the mix without it dominating.
During the verse, we've got a couple of different guitars. These are both recorded with my Strat. We've got one which is just a kind of percussive, constant eighth-note guitar, and on the other side we have these kind of held chords that are shimmery, and I use the tremolo bar a little bit to give them a bit of extra movement. And they have quite similar treatments. I mean, the right-hand guitar has very little going on on it—the kind of percussive guitar—and the left one has an echo applied to it. But let's just, let's dive in and listen to the percussive guitar, and we'll see what I've done to treat that. So I'll bypass the processing. You know, I've been talking a lot about these notches, and here's one appearing in a very different frequency range to where I typically do them. What I found here is because I'm palm-muting that one note—the root note E—so much in this riff, it's a little bit overbearing. So I actually went and found the frequency of that note and cut it. It seems a little bit counterintuitive to cut the fundamental tone of the note that you're playing, but by doing that it actually allows the rest of the kind of percussive character of the sound to come through, and you don't get this kind of bogged-down sound with too much of that low-end there. Again, you might be tempted to take that out with a much wider cut and end up with a sound that perhaps lacks a bit of character. So it's kind of nice to take care of that with a narrow notched filter. I'm also using quite a steep high-pass filter here, set at 78 Hz, just to prevent any extra low-end from being there. So let me just show you—I'll only engage and disengage the fundamental tone notch just so you can hear what's going on. [Music] So it really cleans up the part; makes it sound a lot tighter.
After that, I've engaged a little bit of compression with this amazing de-esser clone made by, by Slate—again, Slate Digital. I've got it set fairly moderately, with kind of medium attack and release settings and a 3:1 ratio, and it's just kind of there to just congeal that sound a little bit more. It doesn't need much dynamic control, but I like the sound of the compressor on this; it's giving a little bit of extra, extra vibe, I guess. I'm not using it on full mix though. So let me show you what it sounds like first without, and then I'll engage the compressor, and you can hear—I love that little kind of that wetness that you get on the attack with the compressor engaged; it's really nice, and and that was all I really felt was necessary to do on that guitar. On the other side, I've got, you know, a similar high-pass filter, a bit higher up this time, and trimming up a bit of low-mids as well as notching out a kind of resonant note up at like the 2.3K area. And I'm also following up with a really long, very analog-sounding delay with a lot of modulation on it, just to fill the gaps in between the chords. So I'll show you what that sounds like. [Music] If I were to bypass the EQ and delay, it would sound like this. So you can hear it does have quite a lot of kind of low-end junk in there, so you can see why I high-passed it so much. And the delay just adds a really beautiful sustain to everything happening there. [Music]
So underneath the verse section, we have a kind of atmospheric pad synth sound, and that, I wanted to use this to give an extra sense of spaciousness and atmosphere. But what you often get lost sometimes with pad synthesis is you get these very strong tones ringing through. So as with almost everything else in this mix, I've found it necessary to cut the specific fundamental tones out of the synth. The idea here is I don't really need much note there—it's ringing in key—but more than anything, I just want this kind of breathy atmosphere around what I'm hearing. Along with that, I'm also cutting some of the extreme high end, just because sometimes synths contain like insane amounts of top-end content which can sit a little bit strangely alongside more traditional elements. And I'm also giving it a little bit of extra bite. Let me show you what this sounds like with no processing, and then I'll show you what I've done EQ-wise. [Music] I'll show you exactly how much I'm cutting those notes. You might have noticed it—it drastically reduces the amount of, of like, I guess, melody note that you can hear in the pad. [Music] [Music] As I mentioned before, these notch filters are absolutely crucial to being able to find a good level for this track in the mix. You know, by taking out those fundamental frequencies, I can now make that pad a lot louder in the mix and therefore actually hear the cool atmospheric elements that it's got. But there's one more thing which I find with a lot of these kinds of very static pad synths, that they can really fill the middle of the mix, right where you've got your bass guitar, your vocal, your snare drum—all these really important elements are fighting for that space. So there's this move that I keep coming back to, and it's really simple; I use it on synths; I use it sometimes on reverbs for...
I don't know, guitars, vocals, anything that's stereo and doesn't—I don't want it to conflict with what's happening in the middle of the mix. You can use this little technique; this is a good one. I put an EQ—it could be any—into mid/side mode, so you can EQ the mono part of the signal and then the side part signal separately. And I just do a big old cut into the low-mid range. I'm not even particularly specific about it, but doing that immediately makes whatever it is you're cueing sound very wide and also immediately means it's not going to be fighting with those really crucial other kind of low-mid frequencies from the centrally panned elements in the mix. So let me show you what this sounds like without and with. And this is such a great alternative to like a stereo widener plug-in that you might have, or you know, can kind of pitch-shifting techniques for sterilizing stuff. I think this is a really, very neat technique that served me very well in recent times. So instantly it just kind of fills the side speakers and declutters what's happening in the middle of the mix. Oh, that's a good one, too.
No, I think vocals are one of the hardest parts of a mix to have like a set approach on. I mean, there's always going to be things that you do similarly every time, but there's so much that's dependent on the specific voice of the singer and also what kind of what range they're using and what dynamic they're singing in. So this track that we're using for the master class has both a very aggressively sung, kind of almost screaming chorus and then vocals got a very softly sung verse. And those two things require quite a different approach because the human voice generates very different frequencies. You know, the harder you sing, typically, the more treble-heavy, the more presence your voice will have. When you sing softly, you can always have way too much body to it. So there's definitely a lot of extra precautions that need to be taken when you're dealing with vocals. And you know, hear the vocals have been presented beautifully; Robin did an amazing job of singing, editing, and presenting these vocals for the song. But if he hadn't done so—and I say if—these verse vocals, which are sung very softly, were grouped on the same channels as our pitch-screamed leads, you might well want to separate them onto new channels so that you can process them differently.
Let's get stuck into the strongest part of the song, the chorus, and we'll look at the techniques which I've used here to compress and EQ and process the the main pitch lead. So Robin tracked his voice with a little bit of compression on the way in, and you can see, even just visually, that these are quite nice sausages of audio; they're not super dynamic and all over the place. Now, part of that's also that he's a really good singer—that's great—and maintaining his volume, controlling the volume of his voice, but a little bit of compression on the way in is often my preference, just so that the the voice is not completely raw once I start mixing it. He's provided me with a main voice and then a double and a triple, so he's recorded the same part three times over and has lined them all up, so it sounds like one big voice essentially. He's also given me a low octave voice, so the—is the same part as he's singing, but recorded an octave lower, and that's something which is just going to get softly blended in underneath the main voice. So let's take a listen to the main voice in the chorus, and you'll hear that there's quite a lot of distortion on it already—partly just as he's overloading the capsule of the microphone and partly the way that his voice is saturating as he sings the part.
At the bottom of my cupboard and frees up. Now you can hear there's a lot of proximity effect in that microphone; you're getting a lot of that kind of lower-mid build-up—kind of sounds like he's singing through cotton wool. So for me, I think with voice, with vocals, I found the best approach is to stick to this subtractive EQ methodology, and you can see my moves that I've done here on the on the voice—a very much cutting low-end, cutting the lower mids, focusing on this kind of more radio-speaker-esque frequency around 1k and cutting some of that, cutting again up in the more kind of presence peaks. Easy, I'm just cutting away loads of the voice to hopefully leave something that sounds a lot more balanced. So I'll show you what it sounds like without and then with, so you can hear that's really cleared his voice up a lot; it's kind of decongested it a lot. And for me, that problem frequencies often are in these areas, you know, your three to five hundred, seven to 1k, or maybe a little bit more like here, then you're two to three k area—those are the kind of typical areas where I found that that really loud vocals often need to be tamed and surfing like this low-mid area. Some English when we get to the quieter vocals is going to need a lot of attention to get right.
Moving on from the EQ, I've chosen to hit it with some fairly heavy FET-style compression from Slate Digital—again using this kind of blue stripe-esque compressor with attack and release on full. Now, people use all sorts of different settings on these compressors for, for vocals. I like to have the attack all the way up when I use an 1176 or a derivation of it. I find that if you turn the attack back, yes, you get this nice percussive element to the voice, but I also find that you tend to get quite a lot of s's being exaggerated, and this you'll see through the compression stages on the voice. One of the things I'm really trying to minimize is the amount of s's that's coming through or make it sound natural without having to resort to kind of unnatural de-essing. So keeping the attack on full, and this kind of compressor for me does a really good job of making sure the voice sounds very smooth. I'll show you what that's doing now by playing it to you first without compression, then including the compressor, so you can hear that that's really gluing his voice in. I started to make the voice sound very non-dynamic, which is a good thing because he's got to sing up against these very distorted guitars that are just basically flatline as far as volume goes.
Next up, I've got another round of compression, but this is a really important one to my process. Again, I'm using a FabFilter plugin, the Pro-C, to order in the vocal mode. The really cool thing with this plug-in is that you can EQ the sidechain, so this is what the compressor is acting on—it's not actually hearing the signal that you're hearing. But what I love to do with this compressor on, on vocals is to heavily emphasize the high frequency. So you can see here I'm boosting 15 dB at 8 or 9k. The reason why I'm doing this is because when you compress in more typical ways with normal compressors, you find that the esses can get really out of control going into the compressor. These s's are generally very quiet when you see them on the waveform, so therefore they don't get turned down in volume alongside the rest of the the sounds that the vocalist is making, and they come out of the other side way louder by comparison to the rest of the vocal. So that's why you end up needing to use de-essers and get creative to them, cut back those s's. However, if you've got a plugin like this where you can heavily emphasize those high frequencies, you can end up with a compression where the esses are treated very similarly to the rest of the vocal sound and therefore don't come out sounding way out of whack in comparison. So this is a really important part of my vocal processing. Today, I really love this trick; it generally means you don't need to use a de-esser on your vocals, which I really like. I hate hearing kind of lispy-sounding vocals. I'll show you what this sounds like. I'm also using this just for general compression. For me, with an aggressively sung vocal like this, it's just about getting as much compression as you can possibly get onto there in order that sounds completely pinned and really aggressive. So this is also giving you quite a big volume booster—makes it more difficult to hear—but essentially this is giving us just a next level of flatlining on the vocal, but hopefully you can hear the percussive t's and the s's are not getting over-emphasized when they come out this compressor, and that's really due to this this high-end boost in the sidechain.
So that's kind of where I end up on my individual track processing. So there's some compression on the way in, got something here, then we've got two more stages of compression, and as we'll see, there's gonna be another stage of compression on top of that, all to try and just get this vocal to be as non-dynamic as possible and to sound really full and upfront all the time. However, there's also a bit of ambience needed to get this vocal to sound the way I like to hear a vocal, you know, so it's got some sustain and body in the mix. So I'm gonna start with a little bit of delay—got a dual echo here with two quarter-note delay, so we've got kind of a wide delay, but it's the same delay time on both sides. You can see here that I'm cutting a lot of low and top-end out of this again, just so that it's not interacting too much with the original vocal and just kind of providing a bed of sustain in between the words. Something which I'm also doing at the moment is I'm using a very similar technique to what I used on the verse pad synth to cut out that kind of low-mid build-up on the delay trails to on the voice. So I'm again got a mid-cut here going on at 450 Hertz more or less, but only on the mid channel—that's gonna push this delay further out to the sides. I mean that the voice doesn't end up kind of getting masked by its own delays.
Moving on to reverb, I found this preset in Logic's Space Designer a while ago, and it served me really well across a load of mixes; it's called Hard Ambience, and it's just a really good sound for a reverb on a voice for me. It's a—it's not particularly long; it's not like really lush-sounding, but it does a good job there, giving a sense of ambience and length to the voice without being kind of too dark and muddying up other things and not being too zingy either. So here's what the voice sounds like—I boost the the reverb up a little bit so you can really hear it. I really like that it's kind of almost got like a gated feel to it—like it's there and then it drops away really quickly. So I like the sound of this reverb, but little goes a long way. I try not to rely too heavily on the reverb; it can make things sound a little bit dingy if you're not careful. So there, vocal with the EQ, compression, delay, and reverb—it sounds like this. And then we have the same thing on the vocal double and triple, just that they are a lot quieter than the mix. I think I've got them about 10 dB quieter, and they're panned out a little bit to the side. So when you hear the three performances of the same part together, it sounds like this. So you're just getting that really wide, full, and just great commercial vocal sound that way.
I also like to apply a bit of processing to the submix of these chorus vocals. So I found that upon putting these vocals into the mix, it sounded a little bit thin and edgy, so I found myself cutting just a little bit around—well, I guess about 3k—boosting a little bit of body back in, but these are really small moves, you know, -1.5 dB, +1 dB, just to smooth the vocal a little bit. I also found a couple of frequencies that needed notching in his voice—it's really normal; the vocalist might have these kind of very strong resonant frequencies that are related to the dimensions of their throat when they're singing, and particularly when they're pushing really hard, they can get quite abrasive. So I found a couple here that I wanted to cut. I'll show you—I'll show you what this vocal sounds like without and then with with those notch filters. It's a subtle difference, and it's not on every word. If I was doing a particularly conscientious job, I might go and automate those cuts onto just the points where they're necessary, but I don't feel I'm particularly doing anything bad to the vocal by leaving them on all the time. Finally, on the chorus vocal bus, I'm using another instance of our smash-and-grab compressor, again set to the overhead mode—which that's a good overall mode if you want to try using this compressor on non-drum instruments. So here I'm going to use a little bit on the voice; this is going to give me a little bit of extra control just volume-wise because we've got a couple of extra vocals popping in and out, and I don't want the voice volume to be changing around too much as that happens, nor do I want to be having to automate to an extreme. So I'm doing some fairly heavy compression, but then I'm blending back a bit of the dry signal in so it doesn't actually sound like a kind of crazy compressed vocal. I never dated it like I do now, and that's those chorus vocals taken care of. [Music] [Applause]
Moving to the verses, I mentioned a little bit of a different treatment required here; it's not worlds—worlds different, but you can see I'm cutting a load of low-end out of the voice here because he's just singing it at a much softer dynamic. So I'll show you what the the raw vocal sounds like. All bottled up thoughts leave me out of touch. By applying this EQ here, we're gonna get rid of a lot of that kind of built-up proximity effect. All bottled up thoughts leave me out of touch. This time, I'm only gonna use one stage of compression, and it's gonna be the one with our high-frequency emphasis; it's a slightly different setting—this is all just kind of done on the fly—but similarly, I'm using the vocal mode of this Pro-C with fast attack and release settings. I'll show you what that sounds like. All bottled up thoughts leave me out of touch, searching in what brings them to the surface. So I mean, I'm compressing pretty hard here; we're looking at like 15 dB of compression—times, sometimes a little bit more—but I really like how this compressor sounds—surprisingly transparent given that it's got a kind of variable ratio—that's just very well designed; this compressor—it sounds great on voice. So that's doing a very good job of leveling out the vocal without making it sound too constricted. But what I really like to do on the softest—on vocals—is to use a multiband compressor as well, which is going to kick in when there's kind of too much body on the voice, but also not mean that you have to EQ out too much low-mids from the vocals and end up with some notes sounding thin and another sounding just about right. So here I've got the multiband compressor set to be on most of the time, but it's really going to ensure that the low-mids are just kept in check all the time. All bottled up thoughts leave me out of touch, searching in what brings them to the surface. Using the same delays and reverbs that are used on the chorus vocals, just in slightly different quantities, to add a bit of space. So that leaves us with this sound. All bottled up thoughts leave me out of touch, searching…
We're starting to get close to a really good vocal sound for this. I did just find that within the civil answers there are some pretty harsh frequencies going on, and these are the same frequencies that I notched out of the chorus vocals. They go quite a long way on this verse to make sure that the kind of sure sounds and the all of those sounds don't have too much kind of peaky information in there in the way of very specific resonant frequencies. And I also found that a little bit more low-mid control is necessary—like a kind of 500 Hertz cut—and with all of that in place, we kind of end up with the final vocal sound. During the verse… up thoughts leave me out of touch, searching in what brings them to the surface. As far as the double goes, it's very subtle in this part; I don't want a very obviously doubled vocal sound because I want it to sound quite intimate. However, I'm using the same processing, and I'm using this SoundToys MicroShift plug-in to stereo-wise it a little bit, so it kind of sits out to the sides of the main vocal a little bit. I like this technique; when you've got a single double, it can be really nice to give the vocal a bit more width when you don't have a triple like we did in the chorus. So—so here's how the double sounds with the stereo widener applied to it. Searching in what brings them to the surface. If I make it too loud, the double's kind of chorusing effect is gonna make it sound a little bit out of tune, so it has to be really quite quiet in the mix. I think I had it about 12 dB quieter than the main vocal down.
I also set up a little channel here with a vocal delay throw on it, and what that is is a channel that just has a nice signal chain for a very airy-sounding delay throw. Essentially, we've got our main delay here, courtesy of EchoBoy—a Memory Man kind of very analogue, rolled-off sounding delay with quite a lot of modulation to it—but I'm taking all the low-end out of it that I can with the plugin, and I've left it with a load of feedback—and then throwing that into a reverb—this great reverb from Valhalla, I called the Vintage Verb—which again I'm cutting low-end from. With these elements, it's really important to keep, you know, the low-end and low-mid out so that these sounds can just be airy and sit on top of the music instead of kind of blocking down the main vocal sound. So we've got delay into a reverb, and then using a compressor on it because what that will mean is that there isn't as much of a drop-off in volume through the delay repeats. So the initial delay repeat, which is gonna be the loudest, is actually gonna get compressed down in volume, and then the compressor is gradually going to compress less and less on each sequential repeat, and you end up with a much more even sound—that's a nice little technique. I think compressing your delays and reverbs can be a really cool sound. And then finally, I'm cutting out even more low-end and low-mids to just make this a very airy sound. What you can do once you've got that set up is you just copy—you just chop out a little bit from your lead vocal and drag that bit of audio up onto your delay track, and you've got a ready-made vocal throw, and it means you don't have to clutter up your vocal with a like a reverb or delay that's very long—that's just there all the time; you don't have to write all the stuff in in automation. I find this to be a really good workflow for creating these kinds of delay throws. I'm sure what it sounds like just on its own and then what it sounds like in context. Bulldog thoughts leave me out of touch, searching in what brings them to the surface. I also use the same channel during the chorus just to highlight—I think there's only one word in each repeat—that's the chorus vocal you can see up here, so you can hear that really long trail on there on the word nerve. There's a nice way of adding a little bit of difference onto certain notes and just highlighting them.
Or the final thing to look at would be the harmony vocals, and instead of treating each of these individually, I decided instead to do as much like it on the bus processing. Now, Robin actually bounced down these harmonies, which are made up of many different takes, in two stereo files that are kind of pre-mixed in volume, which makes things quite a lot easier for me as a mix engineer. I like it when clients do that. So just as one example, here's harmony one during the chorus, and it's like a held-note version of that that fills in the gap, and all of these harmonize with the main chorus vocals. So in terms of EQ on harmony vocals, I think it's really important to make sure they…
Don't have low end that's getting in the way of the other voices, particularly here of high pasture about 150 Hertz. But I feel I could have probably even been more aggressive. We don't need to hear the low end of those those notes. The only things would be japax on those held harmonies. It would be nice that they don't sound too thin since they're kind of standing on their own rather than being sung at the same time. So cousins some some low mids at the same time they're here.
I've gone for a lot more multiband compression than I typically do on individual tracks. And the reason for this is because we've got lots of different types of vocals all going through the same bus. And this multiband compressor is basically going to make sure that each range of the of the vocal is being maintained well. So we've got one band looking at your kind of hundred up to 400 Hertz. Got another band looking between that 400 up to I guess about 1.5 K. And then from there on up to about 6 K or so. And all of these are set to be active but most of the time. So they're basically just trimming that though and keeping it very much in place. I'll show you actually what the vocal sounds like with no processing and then engage the EQ and multiband comp. I found of my confidence Reserve. I dunno. So you can see that a hue and multiband comp really neat and things up.
There moving on I've got a bit of compression. I'm using my trusty pro C to hear similar deal with the high-frequency emphasis. So essentially what we have here is two mono compressors active instead of one stereo compressor, which is really good because you've got loads of different vocals panned all over the place. And if the right vocal pops out is going to compress down the left vocal if it's set and your kind of standard stereo Lynx mode. So by keeping the two sides separate, the two sides can compress at just the amount that's needed for each side to maintain a nice even compressed sound. So compression applied to all of the harmony vocals together sounds like this. I found my confidence reserve. That instantly has made them way more punchy and up front.
I've used to gain plug-in here just to cut some volume. You have seen a couple of other gain plugins being used in my in my mix in general just to trim the relative gains to where I want them. And I'm doing here actually the same move that I've been doing on the pads and on the vocal delays where basically I'm cutting out that low mid content just in the middle. These vocals were all generally pretty hard panned on the harmony vocals, but there's still some kind of low frequencies in the middle that are going to be eating into the space that's being taken up by the the lead vocals. So that's a cool move. And then this guy here is actually affecting all of this spectrum. It's not just in the middle. And I'm using that to cut around 1k to just give a softer sound to the backing vocals. So so I'll start by playing this without this mid/side eq engaged and then with the mid sighting cue engaged again while it's playing. I found mine define confident reserved. So you can hear that's instantly giving us a super stereo kind of vibe to the backing vocals. And I'm rounding all that out with a little bit of delay and reverb just to set this back into the mix a little bit alongside the rest of the vocals and our final harmony vocal sound.
So listening just from kind of midway through the verse into the chorus with the vocals so loud so you can hear what I'm doing there. Let's just take a little quick listen. Trace. I know one day that. So I've taken me through my process of how I mix this track. Let's just take a little listen so you can he how I mix the final product and how all these different elements fit together. [Applause] [Music] [Applause] [Music] [Music] brings them to the surface. Tres [Music] We hope that you've enjoyed this mix walkthrough and who knows you might even be doing another one sometime soon. But in the meantime for more information head over to get good drums.com