Dance Pop Tutorial Day 5: Processing Pop Vocals + Auto Tuning [7 Day Song]
http://www.revisewithrachie.com – That’s Rachie’s Psychology revision website!
http://www.8notes.com/resources/notefinders/piano_chords.asp – That scale finder I mentioned in the video.
Hi! I’m Dave from boyinaband.com and welcome to day 5 of the 7 day song on Dance Pop, courtesy of Propellerhead Software’s Music Making Month.
Yesterday we finished off the synths for our Dance Pop epic, and today we’re working on the all-important vocals. As you’ll notice, I’m in Record 1.5 instead of Reason 5 now – Reason isn’t ideal for vocal recording, so either get rewiring or grab yourself a copy of Record which integrates perfectly with Reason – as you can see, all the reason devices are still in tact, and I’ve recorded all the vocals in here as well.
We’ll take a look at the processing on the vocals, autotuning, techniques for vocal layering, harmonies, stuttering and the hooks themselves.
So first you need to find a vocalist. For my vocalist, I woke up next to her and figured “yeah, she’ll do”, which, as I understand it, is the usual method a record label executive will use to pick a pop star.
But yeah, thanks to Rachie for helping out with the vocals and lyrics on this one – check out her website www.revisewithrachie.com if you happen to be interested in learning psychology, since that what she teaches on her site! Link for that below.
I recorded the vocals into record with my Rode NT-1 mic through a Focusrite Saffire Pro. These won’t sound as clear and crisp as the pro recordings since I don’t have a £4000 microphone, but you can get a lot of the way there by applying some effects to the vocal.
So starting with the processing – the lead chorus vocal I’ve got is a good example of a processed pop vocal. If we solo it out, then take a listen to it clean… yeah, since she helped with the words, lyrically it’s a cross between a sexy dance pop chick being overly confident, and a schizophrenic person coming to terms with their debilitating mental illness. This is what happens when you get a Psychology student to help write a pop song.
Incidentally, I think it’s pretty cool I’ve managed to fit two separate debilitating mental illnesses into this 7 day song.
Where was I? Ah yes, processing. To start, I’ve got compression – this is vital on vocals, especially in pop music; it just smooths out the levels, so for instance when she says “To me to me” it’s considerably louder than the “there yeah you” part at the end of the line. This allows for a consistent volume so you can hear everything nice and clearly.
So I have a decent input gain to bring up the initial volume, try to have the recording as loud as you can get without peaking initially, but the input gain just helps bring that up, the threshold is down to -28dB so it’s quite a pronounced compression, the 4:1 ratio is also quite high which brings down peaks nicely, but not all the peaks – I’ve left the attack up a little bit so the initial sounds of the vocal can have a bit more presence. for instance, the “m” of “me” sounds quite nice hitting nice and powerfully.
The release is right down though so everything under the threshold gets brought up to the threshold nice and quickly, just smoothing out the quiet bits to a more consistent level with those loud ones.
That’s the compressor, next I’ve got the autotuner – this one isn’t set to hard tuning, which is that T-pain and Cher style vocal effect, I’ll talk about that later, but just helps tidy things up very slightly. I’ve set the root note to D, and the scale to natural minor, which is the key of the song. To find out the key your song is in, the root note is the note you feel you’re coming back to a lot and the scale itself can be found by looking at all the notes you’ve used and going to a website like 8notes.com or something and using their scale finder to find the scale that uses those notes. You’ll eventually get a feel for whether something sounds major or minor and most pop songs will be one or the other.
So the settings are correction speed and preserve expression on around middle for both – this means that the notes are being tuned quite fast after being recognised and a bit of the original sound is being kept so it doesn’t sound like a robot.
Next up is an EQ – again, this is really important. I’ve brought up the low end, around 200hz, which gives a lot more power to the vocal, then cutting the mids around 500hz to remove the boxiness of the sound, just makes it sound clearer, then slightly boosting the hi shelf to add that last bit of clarity – just listen to the difference before and after each of these are added.
Notice that the high end, while adding airiness also adds a lot of harshness with the sibiliance – sibilance is the esses and tees that people say, so just notching them out with a little 8khz-ish drop prevents that harshness from going over the top.
Lastly, that maximizer brings everything up to a high level with the limiting – I’ve applied a slow attack so the compressor’s attack can be maintained, but the soft clip just stops it from getting too loud, instead applying that nice warm distortion to any bits that peak at the start there. This gives us something that sounds like this. So without any effects… then with them. You can hear the difference.
On top of this, in the channels I’ve applied send effects for reverb and delay. In true pop style, the delay has filtering applied, I’ve taken out anything below 400hz with the stereo imager to make it less messy in the low frequencies, and widened it as well with a reasonable pan to make it sit just off center so everything isn’t squashed in one section of the stereo field.
The reverb is really subtle – if you make it heavier you’ll begin to muffle the original vocal, which can be good for backing Ah~ kind of vocals, but for the lead you’ll want it to be much more focal.
Phew! So that’s the processing, join me in part 2 for a closer look at auto-tuning!
So this wouldn’t be modern dance pop without at least one example of hard-tuning with an auto tune unit.
Neptune is great at doing this, so let’s pop over to the verse and see what we’ve got.
As you can see here, I’ve given the neptune instance inside this vocal channel a track of its own, with notes in it that define what pitch to play the vocal at.
This allows for some really cool and specific hardtuning. The double track can be exactly the same pitch as the lead vocal. To set this up, just right click neptune and near the bottom there’s “Create track for Neptune”.
I’ve also edited the automation for the correction speed so it is only hard tuned when I want it to be. Hardtuning is achieved by having an instant correction speed and none of the original expression. Check it out.
Nice and processed, just the way I like it! Another thing to mention when talking about how to autotune is formant – by turning on the formant option here in Neptune, we automatically correct for the change in voice tone as well as pitch.
Ever heard one of those chipmunk style vocals? They are caused by pitching up vocals without correcting the formant. Take a listen as I take up the shift. There’s the chipmunk-y effect. Taking it down gives that Britney Spears-y deep vocal that she has on a few of her more recent songs. Or rather that Max Martin has put on her more recent songs.
Semantics aside, this is not on all autotune software, so be careful if you’re tuning more than a few semitones on autotune devices that don’t have that option. Record users though, you can put your legs up, have a cup of tea and autotune the living daylights out of your vocalists without worrying about chipmunkery.
That said, if you go too far the voice ends up sounding really thin and unnatural in a different way. Kinda like an old person wheezing. Take a listen if I play a low note on my midi keyboard while playing the vocal.
Here I’ve also turned on a filtered delay on the last word, always a cool effect – with a nice high feedback so the delay lasts the length of the gap here. This is another one of those really useful things about auxilliary send effects as opposed to ones applied directly to the sounds, they’re much easier to apply timed effects to like this.
Lastly for the autotune stuff, if we head to the intro, I did a customary heavily processed vocal part for the producer, in this case I just said “You gotta” and pitched it down with neptune’s transpose option – simply lowering the vocal pitch without the formant to make it retain it’s human sound.
And I also applied a chord with Neptune on the “Take Me” vocal, effectively turning the device into a vocoder, since I removed the pitched signal entirely.
Okay, now join me in part 3 of the vocal day where I’ll finish off by talking about how I layer the vocals, the harmonies I’ve used and the stuttering.
Layering vocals is another one of those things that you really need to do to get the big, professional pop sound. For the choruses, I’ve got the main lead synth I mentioned, along with two more vocal takes panned left and right fully to really give a wide sound. The vocals left and right are slightly quieter than the lead – they are only there to fill it out.
You have to be very careful with multi-tracked vocals like this, especially if they have t or s sounds since they stick out like a sore thumb when they are out of time with each other and just end up sounding messy.
It’s good to vary this – I’ve not used it in the verse or prechorus so it make more of an impact when it’s used in the chorus, which should obviously be the big, hard hitting vocal section.
That said, I don’t think I’ve got a section with only one vocal. The verses I’ve got a vocal doing exactly the same as the lead track, slightly quieter. Double tracking like this is a good technique to just thicken up a vocal.
Having them at a similar level gives a much more thick and pronounced effect, more like a gang vocal kind of thing, but if you want to retain the intimacy of a single vocal, just mix it low down and you’ll get a nice noticeable thickening of the voice without it sounding like two people.
Now let’s look at harmonies. Firstly, pop songs are good at having harmonies everywhere, just subtley in the background to thicken up a vocal. The choruses will usually be more prominent, and as you can see here I’ve layered up more harmonies in the final chorus.
This is another typical technique found in Dance Pop – do some critical listening to some chart tunes and the last chorus will generally, apart from the conventional ad libs over the top, have another harmony line as well.
Take a listen to the difference – this extra harmony just makes it thicker. Sometimes the first chorus doesn’t have any harmonies, making the final choruses even more impactful.
The preverses have that subtle harmony in there tucked away, just thickening it up a bit.
You can be quite creative with the notes you use in the harmonies, though most pop music tends to either harmonise in thirds with the occasional minor bit – take a listen to Britney Spears “If you seek amy” for an example of how a quick minor harmony can make for an awesome quick dark feel to an otherwise upbeat pop song – or to use a single note for the majority of the harmony. Particularly in choruses. This just adds a crapload of tension and prettiness to the vocal, making it really powerful.
I decided to come up with a harmony that focused around a semitone in the key for the chorus after I’d done the vocal takes – this is when neptune is really useful. Firstly, you can get away with a lot of weird sounding vocals if they’re used as a subtle harmony, so don’t worry too much if the formant settings make it sound a bit croaky – you won’t hear it that clearly behind the lead vocal anyway. Have a listen to the harmony on its own… and then with the lead. just fills it out and adds that gorgeous tension.
the bridge harmony is much the same as the chorus in this case, since I wanted it to build really powerfully. I’ve introduced more harmonies the second time around to make it more interesting but still memorable. Harmonies are great for that.
Lastly let’s talk about stuttering and the hooks themselves. Dance Pop uses repetition. I can’t stress enough how much Dance Pop uses repetition. You see, it’s really important to remember how much Dance Pop uses repetition. It makes things stick in your head. Stuttering is a relatively modern technique and has the convenient side effect of repeating a section of a vocal.
So a Dance Pop song hook if it’s meant to be catchy will either repeat a melody line quite frequently or a whole vocal phrase. I don’t just mean throughout the song but for instance in the verses here, the p-p-people has stuttering at the start which is memorable, the why why why is repeated and the not at the end has heavy delay, allowing it to be repeated.
I read somewhere you have to say something 7 times before it sticks in your head. I don’t know how much truth there is in that, but by this logic, you should be able to remember the chorus at least if you have at least one repeated phrase in there. For instance, we’ve got the “To me To me” section in the chorus, as well as that repeated melody throughout, and that “Just an illusion” at the end there.
Stuttering is also a good way to repeat something without it coming across as repetitive. Stuttering the “Take me” in the intro here has allowed it to be a memorable vocal, but makes it a 2 bar loop instead of just one bar since the second bar has a quick vocal stutter.
By the way – to stutter, it’s just a case of zooming right in, cutting out part of the vocal and pasting to taste in a rhythm you like. General in dance pop it won’t be vastly complicated. Take a listen to Ke$ha’s “Blow” for an example of stuttering being used to make an otherwise boring chorus really cool and catchy.
I’ll just do a quick example of stuttering this intro vocal in a different way to show you.
Okay! And that’s about it. Vocals are a huge aspect of Pop music and I’ve only scratched the surface with what I’ve said here, but I have covered a lot of the more important things to consider when making a big pop vocal. Different voices react differently to different techniques so don’t expect everything that works on Rachie’s voice to work on yours or your vocalist’s, and vice verse – if you think there’s something here that you don’t think suits Rachie’s voice, try it on your track and it might sound good. Vocals are really hard to generalise since they aren’t usually as replicable as a synth sound!
So join me tomorrow for Day 6, where I’ll discuss how I’ve arranged this track and how I’ve used FX to fill it out, as well as the notes I’ve used in it. See ya!