Get a free year on Tuts+ this month when you purchase a Siteground hosting plan from $3.95/mo
While known for its iconic "wub wub" or "wow wow" bass sounds, Dubstep has developed beyond these two vocalizations. You now hear "eee", "eugh", "yah" and a host of other speech like formants that really start to sound like speech; just through a devilishly deep bass sound.
However the amount of trickery that is sometimes required to get beyond "wub" and "wow" can be daunting and simply not worth it. But what if I told you there was a way you could musically freeform these vocalizations yourself and edit them later? Better still you could perform your music live and improvise these vocalizations in real time? Interested? I thought so!
In this tutorial we will create a template that could be used in any project or easily created in an existing one. We will examine the template from a larger perspective and how it operates logically, and we will also get right down to the nitty gritty parameters of sound design. So if you are looking to make your producing life easier while arguably becoming more musical in the process, take the jump!
Signal Flow and the Template
Before we can dive into the particulars of this setup we first need to understand conceptually how this template will work. At its core, this process involves using a harmonically rich bass synth, a microphone, and a vocoder. The reason we are opting for the vocoder in this template is because of its ease of use and how quickly we could potentially see results. To create an "eee" or "aaa" sound, without changing pitch, your voice has to accentuate different harmonics above the fundamental pitch. If all you had to do were EQ 2 kHz and 3.2 kHz to make an "eee" sound then you wouldn't need the vocoder.
But the correct frequencies to accentuate change depending on the fundamental pitch; this is why we have formant ratios and not formant frequencies. A certain formant is created when you accentuate certain frequencies at a given ratio above the fundamental pitch. While you could theoretically do this with automation and an EQ, the amount of work would crazy. The vocoder however lets us do the same thing quickly and painlessly.
In our setup both the microphone and the bass will pass through the vocoder and the bass synth will act our carrier waveform, and our voice through the microphone will be the modulator waveform. Essentially the carrier signal carries all the audio through the output of the vocoder, but the audio is modulated by our modulator signal before it leaves.
All well and good but what does that mean for us? It means that our bass synth will take on the character of our voice. The way it does this is that the vocoder splits the bass and voice into various frequency bands. As the different bands in the voice signal change amplitude, the corresponding band in the bass signal will change volume as well. Another way to think of a vocoder is that it is a multiband dynamic equalizer.
Without diving into too many more particulars for now, lets setup our basic template in your DAW of choice; I will be using FL Studio. Here is a diagram that represents how the audio flow needs to work in the DAW regardless of how it may look on the surface...
As you can see we need our vocoder to be on our bass track since almost all vocoders receive their carrier waveform via the main input. In addition, the vocal track (our modulator) enters the vocoder via sidechain since it is controlling the carrier but not actually audible. Also take note that our vocal track does not enter the main outputs. The reason for this is we do not want to hear our voice at the same time we are vocoding the bass as that would ruin the whole point of the vocoding. Some DAWs may require you to mute the output of the voice while others may simply need to fader turned down so consult your DAWs internal routing options for more on how to do this.
For FL Studio users side chaining can get a little funky at times so here is my template that I have created for your reference. Notice how the sidechain goes to the bass track but not the master output. In addition, FL studio has a quirk where the gain knob on the sidechain has to be all the way down, otherwise you will here it in that track too! If you need more assistance please see my FL Studio Mixer Basix tutorial.
As you can see the base template conceptually is not too difficult once you do it and have a visual aid. However, the tricky part comes in sound design. Without the right kind of bass sound and the proper settings on the vocoder you will get very minimal effect and end up with a boring sound that doesn't at all resemble speech. Lets now take a look at the core aspect of our sound, the synth bass.
The Synth Bass Tone
Dubstep is known for its harsh grungy basses that jump from the bottom of our hearing range right up into the stratosphere at a moment's notice. Part of the reason for this is because they are very responsive to filtering which of course lets you create the iconic "wub" sound. Since our vocoder is arguably just a glorified filter, we too will need a rich bass sound in order to accurately hear the formant shifting from our voice. I will be using FL Studios Sytrus synthesizer for this part of the tutorial but any semi modular FM synth can do the same thing or get you close. Flexibility is the name of the game here and filtering is not an option since we are filtering later so try to avoid a subtractive synth if you can.
Before we proceed keep in mind if your bass sound is going to the vocoder and there is no modulator you will not hear it! So while designing your bass sound make sure you have turned off the vocoder. Otherwise this next section will get a little tricky to do!
Since we need a harmonically rich sound, a waveform such as a saw or square would be a good place to start but you can use any harmonically rich waveform if your synth allows for others; I will be using a square wave as my bass since I like its slightly hollow sounding quality. Now since Sytrus allows me to add harmonics to waveform I decided to add just a tiny amount of upper octave harmonics to the sound to give it even more bite.
Next we need to thicken up the harmonic spectrum and get a grittier sound. To do this I set my second operator (oscillator for FM synthesis) to a sine wave and had it modulate my square wave by 74%. However I also set the sine operator so it is always an octave lower than the note I am playing. Here is what we have now...
Finally to add some more nasty overtones I created a saw wave on my third operator and had it modulate the sine wave by about 44% which in turn modulates the original square. While it may sound rather harsh by this point, we have more than enough harmonic content to get accurate vocoding once we add our own voice. But without vocoding here is our finished sound...
To Vocoding and Beyond
Now that we have our synth bass it is time we get to the fun part, vocoding! For this section I will be using FL Studio's Vocodex as my vocoder. In theory any vocoder will allow you to modulate your source however different vocoders will have various other controls that allow for greater flexibility. I recommend a vocoder that allows up to 100 bands, bandwidth adjustment, and band order adjustment. Lets quickly go over what these basic controls are...
- Band Count: how many filter bands you have to modulate. The greater the number the greater the clarity.
- Bandwidth: how much the bands do or do not cross over one another. Less is clearer while more adds crunch.
- Order: Roll off sensitivity of the filter with lower being smoother and higher being sharper.
With this in mind let us start to shape our vocoder for a nice thick sound that does not play well with others. The vocoder can indeed change our thick bass sound into something much smoother but that is not anything we want! I recommend the following options to maintain the character of our synth...
- Set the Band Count to 100
- Put the bandwidth around 200% to 300%. This will allow for a nice thick crunch on our bands that isn't too clear. After all it shouldn't completely sound like your voice!
- Finally make sure the Order is set on Order 2 for a good blend of clarity and effect
I made a very basic beat and bass line and then sang different vowels over it so you can hear the difference the vocals makes. Finally I also made a freeform version in which I just winged it! Have a listen...
Cool huh? But if say your voice isn't very deep? Then you probably will be modulating more of the harmonics and not too much fundamental. How do we get around this problem? Pitch shifting!
By simply pitch shifting the vocals down you in turn change how it effects the frequency spectrum in the vocoder. Now thankfully for FL Studio users, Vocodex already has a modulator pitch shift built in, but for those of you who don't any external pitch shifting will work too. Here is what it sounds like when I bring my vocals down half an octave...
Freeform Pitch Shifted
What about taking it up? Dubstep always has those screeching high notes that come around as well. Here is an example with me automating the high end for part of the freeform...
Freeform Pitch Shifted with Automation
What is also cool is that you could have a foot controller and do that pitch shifting live if you wanted to as well.
For those of us who do not want to sit there and program every little vowel change, the vocoding route can be a very viable option. Furthermore, for those of us who want to add some freeform vocalizations this is the only method that really allows for that level of creativity. But this method does still have its drawbacks that you should keep in mind.
- You are at the mercy of your microphone. If your microphone is picking up extraneous noise then that noise will modulate your synth bass. Always try to use the most noise free microphone you can.
- You need to have clear diction in order for this too work. You cannot simply mumble your way through it.
While these are definitely good things to keep in mind, here are a few other tips to improving the quality of your tracks...
- If you need a really clean cut sounding vocals, use a gate on the vocals before it gets to the vocoder. This will keep noise down and give your vocals a little more cut.
- If you need less high end or more low end out of your bass, EQ the vocals first and not the bass. Remember the bass is always effect by the vocals!
Well folks that is everything. I hope you enjoyed this tutorial and I hope to hear some freeform Dubstep from you! Take care!