Create a Modal Synthesizer Polyphonic Synthesizer in Bitwig Studio with Poly-Grid

Tutorial | Jun 24, 2022

In this video, I showed how to create a modal synthesizer polyphonic synthesizer inside Bitwig Studio using the Poly-Grid. I explained how to use a noise burst, an XP filter, and an envelope to create the basics of the synthesizer, as well as how to use transpose and ratios to create partials. I also talked about how to use macros to manipulate reverb, velocity, and attack, as well as how to use a slow machine and delay to add texture and depth. Finally, I discussed how to use a select switch to toggle between different synthesizer settings.

You can watch the Video on Youtube - support me on Patreon

Questions & Answers #

Maybe you dont watch the video, here are some important takeaways:

How can I create a modal synthesizer polyphonic synthesizer inside of Bitwig Studio? #

You can create a modal synthesizer polyphonic synthesizer inside of Bitwig Studio by using the Poly-Grid and the Inlet patch. Start by setting up an oscilloscope and an envelope, then connect the envelope to the oscilloscope. Next, use an XP filter and send the audio signal into it. Switch it to bandpass and double click the frequency knob to set it to C3. Set up a peak limiter to form the basis of the modal synthesizer. Then, use a pitch in module to create different partials. Connect the outputs to a mixer and create a macro to control the resonant value. Finally, switch to polyphonic mode and add an ADSR to keep the voices alive.

How do I use a frequency modulation to shape the sound of my modal synthesizer? #

You can use a frequency modulation to shape the sound of your modal synthesizer by using a transpose module. Set up a ratio of 2:1 for second harmonic, 3:1 for third harmonic, and so on. Then, you can use a macro to control all

Transcription #

This is what im talking about in this video. The text is transcribed by AI, so it might not be perfect. If you find any mistakes, please let me know.
You can also click on the timestamps to jump to the right part of the video, which should be helpful.

[00:00.000] Hey folks, welcome back to another video.
[00:02.500] Today I want to explain as simple and fast as possible how you can create a modal synthesizer
[00:08.960] polyphonic synthesizer inside of Bitwig Studio just with the Poly-Grid.
[00:15.320] And I open up here the Poly-Grid and the inlet patch just gives us an synthesizer here
[00:24.680] with a waste table oscillator and an envelope.
[00:29.160] And we can play it on the keyboard because here the pitch pre-code is active, table this
[00:35.520] because we don't need a waste table oscillator at all.
[00:39.240] So we use an oscilloscope here, which is the slow, and instead of using the audio output
[00:47.280] of this envelope here, where you go audio in, then the volume is shaped and then you
[00:51.760] get the audio out here, right?
[00:53.880] We get the signal output where it outputs basically the shape that you dial in here
[00:59.920] with the knobs as a signal.
[01:01.800] So when I press some keys on the keyboard, you can see we get the signal as a shape as
[01:09.840] a signal out of this envelope here.
[01:12.520] And we use this signal as an audio signal, just a crackle basically because we send the
[01:25.640] speakers basically to hell by using this steep ramp here.
[01:31.640] So then we use an XP filter.
[01:34.360] We send this audio signal into the XP filter and out the audio output here.
[01:39.680] We switch to bandpass to pull and then we double click this frequency knob here and
[01:45.520] it snaps to C3 and because here the pitch pre-code is active and the gate input pre-code
[01:55.120] is active, we can play now on the keyboard.
[02:02.600] We use here a peak limiter, okay, so this is now the basis of our modal synthesizer
[02:13.800] because we have a noise burst sending it into a filter, excite this filter and with the
[02:19.240] resonance we have self-resonance and then we get sound out of it.
[02:27.640] Nice, so now we need to create some partials because this is only the root frequency or
[02:37.160] the fundamental.
[02:39.160] So we go for, let's start with pitch in, hit this here into that and we disable the pre-code
[02:49.640] and we dial in the amount of modulation.
[02:55.280] So now it's basically the same as before but now we have a separate module here for
[02:59.400] the pitch in when we can hook up something in between here, it's still playing, it still
[03:06.400] works, right?
[03:07.880] And in between we hook up here a transpose and then we can start to just, yeah, second
[03:19.680] harmonic, we can dial this in here with the numbers 2, 1.
[03:24.520] So this is basically the ratio.
[03:28.400] You can also use here the ratio module if you want to and just dial in 2, 1, basically
[03:35.240] the same but afterwards I want to modulate probably the transpose at some point with
[03:40.680] some modulators so I'm sticking here with the transpose option and ratio 2 to 1, exactly
[03:50.080] 1 octa fire, nice, 3, 1, 4 and we do 8 partials, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16,
[04:19.960] 18 and 8.
[04:25.320] Now we need to bring all these, all the outputs together and we use some module, basically
[04:33.120] a mixer but without mixing controls, all these outputs in there, easy.
[04:49.520] Then we get the output here, go to nice.
[04:53.920] So now we need some controls to shape actually all devices at once, for instance, for the
[04:59.080] resonant value, we go here for macro, all this resonance, then we modulate with this.
[05:16.320] So now we dial in here modulation, all resonant knobs, then we can use the roll, then we can
[05:33.200] just use this knob to dial in resonance for all filters at once, maybe we can just, some
[05:46.640] notes here, okay this kind of works and sounds already nice like an instrument but now I
[05:58.840] want to switch this to polyphonic mode because we want to press multiple notes at the same
[06:04.000] time, at the moment it's just monophonic and now with the polyphonic mode we have a different
[06:09.400] problem.
[06:10.400] Now the voice is cut off because in polyphonic mode each voice needs to have a feature to
[06:18.520] keep it alive, so Bitwig doesn't know when a voice is dying or needs to be killed.
[06:26.320] Now we have like this pretty cut off sound because all the resonating tails here are
[06:32.880] cut off because the AD is the only envelope in here that keeps the voice alive, so we
[06:40.120] need the second one, go for the AD here and when you open up here to help menu we can
[06:51.040] see we have in the inspector here this effect voice lifetime thing.
[06:55.120] So it's basically audio is flowing through this ADSR coming from this XP, from these
[07:02.240] resonating XP filters here having a long tail outputting audio and as long as audio recognized
[07:09.920] inside this envelope this ADSR keeps the current voice of your pressed key alive so
[07:16.240] you can press multiple keys, right and each voice is kept alive until there is no audio
[07:24.320] coming into this ADSR anymore.
[07:26.480] You can also use this here a bit to shape the sound a bit more, remove the harsh attack
[07:35.160] basically, nice, sounds cool.
[07:41.240] So maybe we can also bring in here another filter, let's use the new one MG here the
[07:48.200] MOOC filter, you can use this maybe to shape the harmonics a bit more, it's kind of changing
[08:01.720] the volume of each partial, right, that is like subjective synthesis, so each partial
[08:07.320] is playing at the same loudness and then you remove basically here top frequencies so
[08:14.120] when I use here the spectrum filter, well let's use two after the filter and the four
[08:24.000] the filter, right, you have all the top harmonics and then you slowly subtract every harmonics
[08:35.520] so this is how a subtractive synthesis basically works and for subtractive synthesis you need
[08:40.760] at least the sound that has overtones harmonics so that's why in most subtractive synthesizers
[08:47.160] there is no sine wave or sine oscillator because sine has no overtones so that's only
[08:53.160] saw and tree and square or pulse wave, right, this is the reason for that.
[09:01.480] So now we have here a bit of shaping the sound, we could actually replicate this bandpass
[09:09.760] by just adding here an attenuator to each partial, it would be the additive synthesis
[09:23.040] method of having a low cut or a low pass, so instead of removing here the harmonics
[09:31.800] with this one and dial down each partial until we end up only with the root or with the fundamental
[09:47.720] frequencies so it's basically the same thing just in different methods, so yeah, removing
[09:55.520] the upper harmonics, this is way easier, right, you make this differently and start so you
[10:07.000] can also shape the inputs first a bit, you can see it changes the tonality a bit, you
[10:18.040] can also instead of using this signal output here we can use a noise module and sending
[10:28.640] this noise burst into these filters has a different feel, different vibe to it, it's
[10:39.280] more like a noisy attack, maybe switches to stereo, pink noise, gets more like a physical
[10:52.040] modeling vibe, it's still cool, maybe switch back here to the signal one, but just by shaping
[11:02.160] the input burst noise that you send in you can change the tonality of this output very
[11:11.840] drastically, you can also try to use a low pass here on this one, it has some kind of
[11:26.360] effect to it, just to keep this in mind, you can use all these parameters to shape the
[11:31.800] sound later on or make it more fluid or more alive by randomly modulating all of the stuff,
[11:40.840] what else can we do, we can change the volume of each partial which has a drastic effect,
[11:45.560] we can change the frequency of each partial, maybe we can do something like using a macro
[11:53.960] here and calling this this free function and then we dial back here each of these, yeah,
[12:09.000] which modulations, maybe a water out, modulate everything just a tad or any of the spectrum
[12:20.240] dial it in by a numbers minus one, one, so yeah, inside Bitwig it's a bit of a hassle,
[12:30.760] you have to press every time you dial in a value, you have to press return, you just
[12:35.840] click to the next option here, so if I dial in minus two and press the next option here,
[12:41.800] as you can see it's reset to the original value, I wish we had something like a tap,
[12:46.360] we could just press tap, so it's a save and move on to the next, but it's not in there,
[12:53.480] it would be nice to have tapped like it works in Windows by default, right, so now that
[13:02.400] we have this, we can dial in here basically with the distribution to remove all the frequency
[13:08.960] modulation or transposing here, it sounds like this, we can move from trash can to steel
[13:26.120] drum or something like this, so you can see that all the frequencies are distributed between
[13:36.920] these partials, it's very important for the sound, so you can come up with different algorithms
[13:44.920] maybe how to shape the frequencies, so maybe call this your algorithm one maybe, and then
[13:52.200] you say well this algorithm is changing some of the frequencies pretty drastically, maybe
[14:02.360] zoom out here a bit more, algorithm one changes maybe the second partial like this and like
[14:12.520] this and this direction and this and that, like come up with different frequency modulations
[14:19.440] for each partial, so let's see how this sounds, you get this type of sound, then you create
[14:29.920] maybe a macro and call this algorithm two, and you don't need to do this randomly, like
[14:38.080] I did here, you can completely go and dive into some literature and read how certain
[14:47.280] frequencies are distributed for different sounds and come up with a more unique kind
[14:56.080] of distribution algorithm, just to give you some ideas for experimentation here, this
[15:06.320] gives more kind of bell sounds, now leave the fundamental completely untouched just
[15:17.600] to have something, have a bass in there, then you can mix up the two algorithms, there you
[15:31.160] can do a hundred of these algorithm macros here right to come up with something interesting
[15:36.880] at the end, hey it's a bell generator, we go up, delay two here, now maybe use the
[16:06.760] macro for the cut off here, because there's a lot of high end content there, you can also
[16:22.160] use velomult which changes the volume, the output volume here of the grid based on your
[16:31.080] velocity, how hard you press the key on the keyboard, we can play very soft or very loud,
[16:42.520] you can also use the velocity to change here the attack time, so we use a modulator, the
[16:50.560] hard I press my key on the keyboard is something I do a lot here, styling in a soft tech, this
[16:58.360] is basically when I play the sound pretty soft on my keyboard, then when I press the
[17:03.880] key harder I make the attack sound pretty hard or pretty fast, right so you can make
[17:19.800] the sound or the synthesizer more expressive, what you also can do is use your maybe expressions
[17:36.440] and use the velocity to change the algorithms just a tad, just slightly right, so every
[17:41.880] time you press the key a bit differently then the sound changes a bit differently, or the
[17:53.280] harder you press the key on the keyboard the more you open up the cut off, this is basically
[18:11.440] all you can do to make it feel more real or more alive, so now that we have this what
[18:20.400] else can we do, so we have here this and that, okay another option would be to maybe use
[18:28.720] the same principle we used here with the bandpass filter, we can apply this to all of the output
[18:36.240] of this one here, so we can insert here another XP like this, which is to bandpass, which
[18:46.120] is to the good frequency or C3, C3, then self resonate with this, so it sounds more a bit
[19:01.040] more muffled, you have a lot of, also a lot of texture on top, but it's still everything
[19:08.640] have to go through the self resonating bandpass filter here, which makes it more musical I
[19:14.560] would say, so also what I also like to do is to play back to 5 and then after all these
[19:38.320] 3 burbs and delays here I put a slow machine like this, slow everything down and then mix
[19:51.000] in the original signal, 40% something like this, then you can play and...
[20:09.680] Okay, something like this and maybe this also needs a bit of pitch modulation at the end,
[20:30.840] not sure if we just put this in here, so I use a delay, you can see when we move this
[20:46.840] delay it changes the pitch slightly, we use this to random mod and I always love to use
[20:55.560] random modulators for the delay modification, I know a lot of synthesizers actually use
[21:04.560] an LFO for that, but then you have always the same up and down movement in the pitch,
[21:10.960] but here it's like a bit more random, so I modulate here, too slow, that's too much,
[21:40.800] fast, what we also can do is use delay or long delay, get the output of this and feed
[22:00.000] it back into our self resonating XP here and modulate the filter frequency, let's see
[22:08.520] how this sounds, kind of like it, maybe we should here put in a select so we can switch
[22:36.880] between the two versions, so we have here button here and call this, I don't know, resonating
[22:54.760] and we have to switch the inputs here, this, we have like the original version from before
[23:09.440] and then the second version is here our XP filter, self resonating, of course we can
[23:19.000] use this here also in the modulator section, call this rest or not, or then we have to
[23:48.000] use here also a macro modified amount, let's see more, okay.
[24:05.560] Here I can come up with a lot of different ideas to tweak this, but I think I explained
[24:32.320] here the basics very well and the rest most of the times that's actually true for everything
[24:38.680] in synthesis, the synthesis method itself is sometimes pretty easy and simple to do
[24:47.120] also in the grid, the hard part is basically to find the right values for each module and
[24:56.320] the sweet spots and how you map these remote controls to something and come up with something
[25:02.640] interesting for the end user, so it's like when you buy a synthesizer you imagine or
[25:10.120] your expectation is that you fiddle around maybe for 10 seconds and then you find a nice
[25:16.800] sound and then you think, wow, this is a nice synthesizer, but it's not about the processing
[25:23.840] and the method you do, it's more about the designer, how they choose or chosen what to
[25:29.840] map to what parameter and how you influence it via the interface, so this is the hard
[25:36.880] part basically of building instruments, at least what I learned in recent years, that
[25:44.400] it's very hard to find sweet spots and make an instrument sound great or feel great for
[25:49.440] the end user that just dials in a bit of here and there and then it sounds good, but you
[25:56.960] can see the possibilities are endless, you can tweak this in all kinds of directions.
[26:02.000] I'm pretty sure when I disable all of this here and put maybe a distortion unit here
[26:07.560] at the end somewhere and dial in the resonator here and play some lower parts, get some drastic
[26:24.920] bass sounds out of this probably, the free warp still, they get percussion sounds out
[26:51.760] of this probably and bass drums and bass sounds and whatever, so it's tweakable in every kind
[27:01.000] of direction.
[27:02.000] So I would say I saved this here as a preset, maybe without the VST, so it's Bitwig only
[27:08.200] and this is, let me see, I think this beta 8 here, beta 8 and I put this on my Patreon
[27:14.720] so you can download this on my Patreon for free.
[27:18.800] You just need to subscribe probably for just one buck, but then, you know, you can just
[27:23.280] download this and have fun with it or you just rebuild it from the ground up.
[27:27.000] I mean, it's not hard to do and just follow the instructions in this video and you're
[27:33.520] good.
[27:34.520] Okay, and that's it.
[27:36.160] Thanks for watching this video, guys.
[27:37.160] Leave a thumbs up if you liked the video.
[27:39.360] Subscribe to the channel, subscribe to Patreon and thanks for watching.
[27:43.600] Until next time.
[27:44.600] Bye.