in the previous episode we've seen a couple of things we could do by stacking up two or more oscillators even though layering is a very important concept in sound design there is so much you can do with two oscillators to create your own sound so today we are going to see that and talk about sound synthesis there are a lot of types of some synthesis so let's start right away with the first one additive synthesis is creating sound by adding harmonics in the sound that can be achieved by stacking oscillators on top of each other just like we did in the last video but most of the time we will use sine waves to do that so we can add harmonics one by one each harmonic you add will change the waveform because the waveform is dependent of the harmonic content and vice versa so changing one will change the other you can do additive synthesis in ableton's operator for example here in this area every vertical bar represents a harmonic of the harmonic series and the height represent their volume if you don't know what the harmonic theory is it might be a good time to watch my old episode about harmonics and overtones that i will link in the description everything is explained there so if you have only the first harmonic the fundamental you will have a sine wave then if you add a harmonic every two you would have more of a square waveform and if you add every harmonic you would begin to have more of a sawtooth waveform what's interesting in additive synthesis is that you can control the volume of each harmonic independently so you could shape the sound with more flexibility [Music] you can also do additive synthesis in serum by clicking this pencil icon in the oscillator section and in this window you will have the same display with all these bars with a visual representation of the waveform you are creating [Music] so [Music] so [Applause] so subtractive synthesis is starting with a sound that is rich in harmonics and then removing some of them using filters there are a lot of different filters to choose from and we'll see what they can do in a dedicated video but for now you can remember that subtractive synthesis is gradually removing harmonics using those filters to give you an example regardless a low pass filter will cut the high frequencies of a sound so it can make it less harsh or more muffled [Music] and a high pass filter will cut the low frequencies of a sound which can be useful to leave some space for other layers in your sound for example [Music] so subtractive synthesis works particularly well when you start with a sound with a lot of harmonics to give you an example of subtractive synthesis in its purest form you can start with white noise which contains all the frequencies and then use filters or an eq to let through only the frequencies you want [Music] in a wavetable synth like serum or ableton's wave tables the oscillator is called a table and it actually contains several waveforms you would then have a knob to cycle through all of them and morph your sound this can be a good way to add motion to your sound to make it more organic varying the waveform in conjunction with the opening of a filter or the drive of a distortion can really make your sound come alive this is interesting because you could have any way for morphing to any waveform so the possibilities are endless [Music] some of this synth like serum even allows you to draw your own waveforms or you can import an audio sample and it will extract the waveforms from it for example if i take this sound and i import it in serum it would give me all these waveforms that i can use in my oscillator [Music] among these waveforms you can do a selection you can erase some of them you can keep just some of them and if you have just a few different waveforms loaded here you can actually like interpolate waveforms between them so you can move smoothly between them to do that it's in the morph section let's try more spectral for example see the first one is in index number one and the second one is in index 256 which means that there are actually 255 waveforms to transition from the one to the other am stands for amplitude modulation and the amplitude of the signal is basically the volume of the sound amplitude modulation means that the volume is modulated the volume changes following another signal so for am synthesis you will actually need two oscillators one is the sound source that provide the sound you can hear it is called the carrier and the other is the one controlling the volume knob of the first that is called the modulator so for example we have an oscillator with a sine waveform that is our base sound the carrier and then we have an lfo which stands for low frequency oscillator that will provide also a sine wave but very slow if we make this lfo control the volume knob of our first oscillator it becomes the modulator at this rate you can hear the sound go louder than quieter along with the lfos oscillations for now it's no big deal but if the lfo then goes faster and faster it starts distorting the sound when the modulator oscillates fast enough it begins to distort the shape of the waveform itself and it begins to add harmonic content to the original signal [Laughter] it will add several harmonics for each harmonics already present in the original sounds so if i add another sine wave in the carrier so we begin with two harmonics you can see a series of harmonics being added for both changing the amplitude of volume of the modulator changes the overall volume of the added harmonics from there you can experiment with different waveforms for the carrier and for the modulator [Music] ah [Music] it's worth noting that you can replace this lfo for a simple oscillator as well an oscillator and an lfo are basically the same thing only the lfo oscillates way slower one last thing about am synthesis have you heard about ring modulators these are devices that can make a sound more metallic that sounds a bit like a bear the name ring modulation comes from the fact that early analog devices that made rin modulation used four diodes linked together in the shape of a ring well ring modulation is essentially the same thing that amplitude modulation the only difference is that with ring modulation only the added harmonics remains and the original frequency of the carrier disappears whereas with amplitude modulation both added harmonics and the original frequencies of the carrier remain apart from that they work in the same way [Music] fm stands for frequency modulation and frequency is basically the pitch of your oscillator so fm synthesis is similar to am synthesis in the way that they both involve a carrier and a modulator but here with fm the modulator will control the pitch of the carrier in the same example than before i will link the lfo to the frequency of the oscillator [Music] you can hear the pitch going up and down along with the lfo and in the same way when the oscillator rates of the lfo goes up it starts distorting the sound of the carrier but it adds a lot more harmonics creating potentially a much richer sound one of the differences with am synthesis is that here with fm when you move the amplitude or volume of the modulator it also affects the harmonic content that is added to the sound not only the overall volume but also the distribution of the harmonics [Applause] [Music] now this is the core principle of fm synthesis it can go a lot deeper for example we can add a third oscillator before this modulator one so its frequency goes up and down here is how it sounds at a slow rate so if i crank up the frequency of this oscillator we can get yet another sound from there you can add other oscillators and connect them in many different ways i'm showing you this because fm synthesis was popularized by the dx7 it was a synth that used the oscillators that you could connect in many different ways and this architecture was so popular and versatile that it's been copied in many synths like in the volca fm for example or the software decks which you can download for free you can do many different sounds with fm synthesis it's really a rabbit hole that goes a very long way so here are a couple of examples from dext library [Music] i will also add wave shaping synthesis which generally starts with a simple waveform such as a sine wave a triangular wave or a sawtooth wave that is then distorted by a wave shaper the wave shaper will basically change the shape of the incoming waveform based on a function and there are a lot of things you can do with it it is a type of distortion so we'll see that more in detail in the episode about distortion and saturation but just to name drop what you can do with it you can do some saturation asymmetric saturation wave folding or phase inversion for example heart sync is not really considered as a type of synthesis i don't think anybody talks about heart sync synthesis but it creates a very particular sound so i wanted to include it here as well it is close to am and fm synthesis in the sense that it requires two oscillators one controlling a parameter of the other and what happens is that every time one oscillator finishes a wave cycle it resets the cycle of the other oscillator that creates a break in the waveform and it creates a distinctive distortion but having one oscillator resetting the phase of the other means several things one the two oscillators will have the same base frequency because the master oscillator will force the slave oscillator's period to reset at the master's frequency so if you change the master oscillator's frequency it will change the pitch and if you change the slave oscillator's frequency it will change the timbre [Music] two you will hear the effect better if the two oscillator are detuned or played two different notes because if they are perfectly in tune and they play the same note then their phases would be perfectly in sync and it won't cause any break in the waveform it would be seamless and three the waveform of the master oscillator doesn't really matter because it is only its frequency that will determine when to reset the period of the slave oscillator usually the frequency of the master oscillator is the frequencies of the notes you play on the keyboard and the frequency of the slave oscillator is either tuned from these notes or it can be set to a constant frequency this really depends on how your synth is built and the options you have on it last thing to keep in mind with heartsync is if the slave oscillator is tuned to a lower frequency than the master oscillator it will be forced to repeat before it completes an entire cycle and if it is tuned to a higher frequency the slave oscillator's waveform will be retriggered after one or a few complete cycles and this can give the impression that two notes are playing at the same time so if you swift the tuning of the slave oscillator you will actually hear kind of a harmonic series [Music] granular synthesis doesn't use an oscillator as a sound source instead it uses an audio sample well it could still be a sample of a sound made with an oscillator but still it uses an audio sample and slices it into a lot of tiny audio bits generally between 1 millisecond and 50 milliseconds each these tiny snippets of sound are called grains these grains can be layered on top of each other played at different speeds different length different volume etc this kind of synthesis is handy to create mellow textures often referred as clouds [Music] but it can do much more there is a free max for life plugin that allows you to do granular synthesis in ableton it's called granulator 2. here you can place your sample let's take this one for example and there you can define the size of the grains where they are reading from move the playing heads while you play change the pitch of the grains randomize these parameters and more check out this plugin it's free and very fun to play with granular synthesis is also used by some built-in functions in ableton like the time stretch the time stretch works in a very similar way in other doors but i don't know them as much as i know ableton so i will use this one for the example so time stretch allows you for example to make a sound longer without changing its pitch go go to do so ableton will cut your sound in several grains and then repeat some of them to make the sound longer you can then change the size of this grain in the warp mode section of your sample by default it's on beat in this mode ableton will make a new grain with every transient it finds it's good for small stretches or if you want to make your sound shorter but for bigger stretchers you might want to try other modes first still in beat mode you can set the grain size to a subdivision of your tempo which is really nice to make a rhythmic texture in sync with your tempo or you can make the grain shorter by clicking on this little arrow to make sure each grain is red only once and by reducing this value it is very nice to make a loop sound snappier for example [Laughter] you can also use the texture mode that allows you to set the size of the grain to tania bits all the complex and complex pro functions which are supposed to be of best fidelity to the original sound but with more extreme stretches it introduces some metallic tones that can be very nice also very last tip for this video to create your own waveform you can also load a sample in a sampler and select a very tiny part of it and play it in loop because this part is played in loop very quickly it will just serve as a new waveform i used to do that a lot to design bass sounds starting with a kick drum sample and these are the eight audio synthesis that i wanted to share with you today additive subtractive am fm wave table wave shaper hearthstone and granular synthesis that was a quick overview of how they work but we'll see them again in this series most of these types of audio synthesis add a lot of harmonic contents they do very rich sounds so it's often a good idea to combine them with filters to tame them a little so next time we'll see the subtractive synthesis a bit more in detail and we'll see the different eqs and filter types you can use to shape those sounds in the meantime take care and i'll see you in the next video so you