Wavetable synthesis is a type of synthesis where an audio’s soundwaves are modelled on the single cycle waveform of a sound source.
That being said, wavetable synthesis is an interesting form of audio synthesis. A lot of really cool and interesting sounds have been engineered using this approach. It’s also perhaps one of the easiest and most simple forms of synthesis, believe it or not, one which sound designers can spend oodles of time having fun experimenting with the limitless possibilities of textures and morphing alien-like tonalities that wavetable synthesis can create.
Read the other articles on audio synthesis:
Introduction to Wavetable Synthesis
As mentioned in the other posts, synthesis is at the heart of modern music making and music production. It’s especially the case concerning electronic music. So knowing how to use a synthesizer instrument will go a long way in your music making. Why is that? you may ask…
Well, image a sound in your head right now. Go ahead, think of a sound. It doesn’t have to be a new sound, but something cool you’ve heard in another production by your favourite musician or producer.
If you need help, listen to the cool synth lead used in this track and get back here (starts at around 45 seconds in).
Now, imagine if you knew the properties of that sound, and were able to re-engineer that sound into a brand new sound?
In other words, imagine you are the Dr Frankenstein of music production, giving sonic life to electrical instruments, making them growl and groan like monstrous synthesizer beasts.
Well, that’s what the power that the knowledge of synthesis will give you. And in particular, wavetable synthesis.
That said, what is wavetable synthesis?
Wavetable Synthesis – The Basics
Wavetable synthesis, also known as “sample-based synthesis,” refers to a type of synthesis where the sound constructed is modelled on the single cycle waveform of a sound source. That sound source can be a note of a natural sound, like a voice, drum kick, etc., or a sound from another synthesizer like an FM synth.
Modelling the single cycle wave pattern of another audio gives the synthesist the power to create natural sounding synth sounds without the need to use many of the control found in other synths, as the harmonics of the sound would be already be built into the harmonic partials of the sound source’s wavetable.
Woah-ha-how, slow down there, buddy… is probably what you’re thinking. Don’t fret, we’ll break it down to easy peasy for you. By the end of this post, you should have a thorough understanding of the basic principles of everything just said. Keep going…
How does a Wavetable Synthesizer work?
To know how a wavetable synthesizer works, you need to know how a synthesizer works in the first place…
Audio Synthesis Essentials
To recap (or perhaps refresh), audio synthesis refers the ability to create sounds synthetically (as opposed to naturally), electrically or digitally.
How this works is that sound is produced via a device called an oscillator. The oscillator creates a vibration of hundreds to thousands of times per second. This vibration leads us to perceive a “pitch” or “note.”
Let’s say you pluck the top open string on a guitar, E. There is a vibration coming from that string of 330 cycles per second. You can actually see this when looking at the string vibrating, as it goes up and down (or oscillate) at a rate of 330 times per second. We call this oscillation Hertz, or Hz for short.
That is essentially what a synthesizer does. It creates an electrical or digital signal telling the speakers in your studio monitors or headphones to oscillate so many times, it Hz.
Ha-ha. Let’s keep going…
Playing with sound
Moving forward, there are are many ways that frequency generated can be manipulated to create all sorts of different sound textures.
For instance, you can modulate the change in frequency of the signal itself, which is what is called frequency modulation synthesis. This produces lots of overlapping harmonics, giving you nice, juicy sounds.
You can also create a raw frequency with lots of harmonics (like a sawtooth wave) and filter out the harmonics which you don’t want, thereby controlling the tone of the sound. That’s called subtractive synthesis.
But if you simply want to model the wave shape of a sound you hear without doing too much manual manipulation, you can tell your synthesizer’s oscillator to create a frequency with the same pattern in the sound produced by any other sound in real life.
This is awesome. Up until now, we’ve only covered methods to mimic sounds from via synthesis that is as close to natural sounds as we can. We haven’t, so far, figured out that we can directly model a sound that occurs naturally in the world.
And we do that using wavetables.
What? Keep reading…
What are Wavetables?
First, let’s talk about the shape of sound…
Waveforms
Going back to how wavetable synthesis works, let’s think about how sound is visualized, first.
We like to think of sound in terms of waves. It helps us to track the “oscillation pattern,” or the back and forth vibration, of a pitch’s frequency. This unique pattern is what gives our sound it’s “timbre.”
Did I say “timber?” No, I said “timbre,” like “tambourine.” Blame it on the French…
Synthesizers normally utilize a limited array of waveforms to create oscillations per second. These are generally sine waves, triangle waves, square waves, and sawtooth waves.
For example, in order to create a sawtooth waveform, imagine a device telling your speaker to push out suddenly, then slowly suck back in, then push out suddenly, then slowly such back in. Speed up that motion to many times per second, and you have a sawtooth waveform sound that sounds like this.
Starting with those waves as the basis for our sound, it is possible to alter them to create all sorts of cool sounds. But usually, it would be quite difficult to get a true “natural” sound that way, as most sounds in nature do not look like a perfect sine, triangle, square, or sawtooth. And certainly, they do not create oscillations that repeat so perfectly and uniformly over time.
So, how do we create the juicy harmonics of, say, a kick drum?
Wavetables
When you look at a perfect sawtooth wave, you will see that the same exact pattern repeats, or loops, over time. Each loop is a cycle. That cycle then can be used as the basis of another sound that you want to create. In other words, that cycle is used as your looping “wavetable.”
What your synthesizer does is takes the digital representation of a wave, let’s say, the single cycle wave of a kick drum, and replicate that pattern over and over.
That same cycle will be used to play higher pitches by repeating the cycle over and over at more rapid rates per second.
For lower pitches, the cycle is repeated more slowly. Recall what was said about the vibration or oscillation of the guitar string again. The faster it vibrates, the higher to pitch, the slower, the lower the note. The same principle applies here with wavetable frequency cycles.
However, a wavetable typically consists of just a single cycle, and there are many different cycles per second that change over time to get the various sounds you hear in a kick drum, from start to end. In many cases, you’d want a sound that changes in tonality from start to finish. Like a living, breathing, Frankensynth!
But how do you get these cool effects in your sound?
Evolving Motions
Waveforms normally change over time. Hardly do you ever hear a constant”beeeeeep” sound on a real instrument or in nature. A guitar sound, for instance, would not have the same exact sound, and hence waveform, when you first pluck the string as to when you hear the note fading away. And the change in sound is not as simple as the decay in sound volume, which is to say, how audible it is to us over time. There is a definite “twang” to the change over time of guitar string’s pluck.
A synthesist, then, would need to take separate wavetables for each moment during the transformation of that sound, from beginning to end. That is called “sampling.” And the result would be called wavetable samples.
With this method, you could have as many samples as you’d like. As you play your pitch on the synthesizer, the first wavetable sample tells the oscillator to create sound waves based on the waveform found in that wavetable. In the guitar example, it would be of the “pluck.” As you hold the key down for the sound to keep playing, you can change that waveform, or wavetable, from the pluck to any other part of the sound of that note after the pluck.
If you recall from our subtractive synthesis post, In order for us to create a fairly realistic guitar pluck sound, we’d need to go through several steps (check it out if you haven’t read the article).
You can also use an LFO (low frequency oscillator) tool to tell the synthesizer to change the wavetable samples from 1 all the way through to 64 (if you’re using 64 wavetable samples, for example), gradually over time, so that you get the same evolution in tone as that of the guitar, or kick drum.
You can also loop through all these samples either by reversing or repeating the wavetable samples as you play your note.
This technique can be used for voice as well, and it’s the alchemy behind some of those synthy vocoder sound you hear in some electronic music.
Properties of Wavetable Synthesis
Just to clarify some of the jargons used…
Waveform – this is simply the shape of the frequency wave created by any sound.
In synthesis, the most popular ones are sawtooth, square, triangle, and sine waves. However, none of these exists naturally or perfectly in the real world. And waveforms hardly repeat so uniformly and exactly. See the FM synthesis post where qw talk more about each wave shape (with examples).
Wavetable – Just to reiterate, a wavetable is simply one cycle of a frequency’s waveform. So on a sawtooth wave, the sharp up, then gradual sweep down, and returning back up, is a single cycle. This can be used as the wavetable for another sound to emulate that sawtooth. But if you’re looking for a natural sound, you will have waveforms with crazier patterns than that for sure.
Samples – this refers to the technique of recording and creating a digital representation of the waveform shapes themselves to construct your wavetables. Samples can be collected from vinyl recordings, from nature, or even from another synthesizer.
You can have many samples because no sound in nature or the real world will provide a uniform sound. Just like the sound of water running, it’s always changing, so you’ll need to have many samples to capture that change over time.
LFO – this refers to low-frequency oscillator. If you have a wavetable with 128 samples, you can manually switch through those samples using a knob on your synthesizer. This would give you a nice, evolving sound.
But that would be tedious. An LFO is what you can use to control that knob. You can define the pattern and speed that you oscillator passes through the wavetables over time, as well as how they repeat through the series of wavetables.
What is Wavetable Synthesis Good for?
In classic synthesis, like FM and subtractive, to create a sound that is rich in harmonics and therefore natural sounding, you’d need a lot of equipment and processing power. You’d have to employ filters, modulators, LFO tools and envelope controls to manipulate your sounds in such a way to get anything natural sounding.
Wavetable synthesis does all of that for you, since a sound source would take care of all those tones and harmonics. Wavetable synthesizers then became easy to use, and cheaper to make because they relied on less power and equipment.
But the true benefit is in the amazing sounds that can be possible. Many of those sounds with evolving, drone-like motions, that seem to palpitate and breathe on their own are created using wavetable synthesis.
You can also get sounds that imitate the human voice to create robot speech or singing. And those juicy and gritty synth basses and leads are often modelled on the harmonics of other sounds like percussions, basses, claps, and so on.
Great Examples of Wavetable Synthesizers
Here are some excellent examples of wavetable synthesis, either in your DAW or as a third-party VSTi.
Included in a Digital Audio Workstation
- Simpler by Ableton
- HALion Sonic by Steinberg
- Orbis Wavetable Synthesizer for Propellerhead Reason by Skrock Music
- Retro Synth by Apple Logic Pro
As a Third-party Plugin
- Omnisphere by Spectrasonics
- Serum by Xfer Records
- Electra2 by Tone2
- One by SynthMaster
Wavetable Synthesis – Final Thoughts
And that is what wavetable synthesis is.
In this post, you’ve hopefully learned by now the theory of wavetable synthesis, and how that theory is made practical in a wavetable synthesizer.
You learned that a wavetable synthesizer requires the use of what are called “wavetables” to create sounds that model the harmonic richness of natural sounds, whether found in nature, or on a real instrument. And it does this without the use of too many types of equipment and techniques.
You learned that wavetables are generated by creating digital samples of a single wave cycle of a sound frequency. And that these samples are repeated over and over to create your pitch.
Also, you learned that in order to create evolving sound waves, like those you hear in a drone or speech type synthesizer, you need multiple wavetable samples which you trigger one after the other either manually, or preferably with an LFO tool.
From now on, when using a wavetable synthesizer, you should be able to more intelligently utilize its sampling and wavetable functions. In particular, when it comes to your sound designing projects, experiments, and productions.
If you enjoyed reading this, and found the information useful, share it with a friend! Also, you can learn more about music production from this online course guide which takes you through many of the courses available.
