Golden Master

Golden REAKTOR VOSIM!!

Well, for all you poor bastards who don’t own a G2 modular, and can’t use the VOSIM patch I made for mine, I have put together a VOSIM patch in Reaktor. Actually, part of the reason I made it was just to implement some features I couldn’t get in the G2, and because I thought it might sound different in reaktor.

part of the GUI for the vosim synth

It does sound different, and, actually, I think it sounds better in Reaktor! Part of the reason (geek speak here) is that the ramp oscillator in Reaktor allows more precision for syncing the waveforms and there are less jitter artifacts. Also, reaktor just sounds good :). I was also able to mess around with a couple tricks that are easier in reaktor than on the nord.

another snap of the GUI

The architecture of this synth: It is a 3 OP VOSIM/FM hybrid. The 3 oscillators can frequency modulate each other, and are all fed into the same VOSIM “burst envelope” which creates the fundamental frequency, and the 3 operator/oscillators are what create the formants.

I found that using a little frequency modulation can add a lot of texture to the vocal sound, or just help in creating a lot of crazy textures.

I’ve also included a fairly comprehensive modulation section. There are only a few presets, but there is really a lot of possibilities in this synth, so enjoy!

I made a simple track to show some sounds from this Reaktor patch. It was just a fun quick demo, but there’s some good textures in there, anyhow. The pad/abstraction sounds are from the VOSIM synth. The “techno” sounding synth and the drums are from the nord g2 modular.

unsavory Associations Mp3 audio

and here is the reaktor patch:

golden vosim reaktor patch

1 Comment

Golden Master, March 17th 2010 | Tags: Audio And Synthesis, fm, formant, frequency modulation, reaktor, vocal, VOSIM
Posted in Audio And Synthesis, microtonal, reaktor, software

VOSIM Synthesis for the Masses

I’m always on the lookout for interesting new ways to get sounds, or new synthesis methods. Especially ones that don’t take a degree an DSP engineering to figure out. VOSIM fits the bill. VOSIM is a kind of formant synthesis where you can control the formant (louder peaks in the frequency spectrum, like in vocals) and fundamental frequency of a tone completely independently, without using any subtractive filters.

Some VOSIM waveforms

Above are some VOSIM waveforms. It’s hard to glean exactly what VOSIM is from the Csound patches and weird academic descriptions online, so I am just giving my approximate definitions and explanations here (please correct me if I’m wrong here, inter-nerds). Basically, VOSIM is a chain of Parabol (or sine) pulses that are “windowed” or enveloped into bursts. There can be a delay between these bursts, or they can immediately follow one another. I found that the delay between bursts did not affect the sound as much as the frequency of the pulse trains, and it was tricky to implement (I did have a working version of a delayed-burst version of VOSIM working to get the waveforms in the image above) so I just made a version with continuous bursts (where one follows immediately after the preceding burst).

Breaking it down...

The diagram above explains the Signal flow of my patch. A sine wave is multiplied by itself to give sine^2. This makes the waveform completely positive instead of bipolar. A sawtooth wave that has only positive values is multiplied with the Sin^2 pulses to give continuous Bursts of pulses. The length of these bursts, determined by the frequency of the sawtooth wave, gives the fundamental frequency of the tone, and the frequency of the sine^2 pulses gives the formant frequency.

sawtooth wave (orange) provides the envelope for the pulses (yellow)

There is one other trick to getting this all working–the sine^2 pulses must be phase-synced with the sawtooth wave so there are no ugly clicking sounds. Any synthesis system with oscillators that can be synced should do this just fine. Make sure that the sawtooth wave is ramping downwards, not upwards, as is often the case in synthesizers.

This is a very interesting synthesis method that can be implemented in just about any digital modular system! I think more people should give it a try. And I think there should be more dialogue online about doing cool stuff like this instead of gazillions of rehashings of classic analogue sounds and “how did so and so get this sound in such and such track.” So, anyone else have some cool ideas?

Here is a video I made to illustrate what the VOSIM waveform looks like, and some of the sounds that can be achieved with a simple VOSIM patch: