KRETS is a modular environment that can be used in any DAW compatible with the AudioUnit or VST format.
The application is based on a set of modules that can be connected in different ways, where focus is on low level audio DSP processing. There are (almost) no ready-made effects to hook into an audio stream, the purpose is for the user to create these more or less from scratch.
By combining mathematical expressions and functions (available as modules) a plethora of audio effects can be created, anything from a compression algorithm to generative soundscapes.
Yes! Although we're in an early development phase, we'd love if you would like to play with it. Maybe you'll find a couple of bugs, maybe you'll invent a killer distortion algorithm by accident. (true story: An accidental algorithm is actually the reason that Klevgrand started making plugins in the first place)
In order to use Krets, you'll need a Klevgrand account. It is very free and lightweight. You’ll also need a Internet connected computer to authorize your copy of KRETS.
This document describes how to use KRETS, a low level modular system for Audio DSP prototyping, research and experimentation. The software KRETS, concept and examples are developed by Klevgränd Produkter AB.
KRETS is a modular environment that can be used in any DAW compatible with the AudioUnit or VST format.
The application is based on a set of modules that can be connected in different ways, and focus is on low level audio DSP processing. There are (almost) no ready-made effects to hook into an audio stream, the purpose is for the user to create these more or less from scratch.
By combining mathematical expressions and functions (available as modules) a plethora of audio effects can be created, anything from a compression algorithm to generative soundscapes.
First time open
When KRETS is opened for the first time, it’s necessary to login. Use the same email and password as on the Klevgrand site.
1 - The canvas
This is the surface that contains all modules. Use left mouse button to drag it around, and right mouse button to add new modules.
2 - Input Module
This module outputs two audio streams from the host (left and right).
3 - Modules (added by the user)
All modules can be connected to each other in different ways. Every module has a number of slots, and these slot positions specifies their kind.
The topmost row (green slots) is for audio input, the ones in the bottom are for audio output. The leftmost are for parameter data input, and the rightmost for parameter data output. It’s perfectly OK (and sometimes necessary) to connect an audio output to a parameter input, or a parameter output to an audio input.
Hovering a slot with the mouse will show additional info about the slot such as its name.
Two modules can be connected by clicking a module´s slot, and drag to the destination module´s slot. To remove a connection, just click a connected slot, move the mouse to the canvas and release.
Modules can be moved by click and drag at its name.
To remove a module, right click it and choose remove.
4 - Output module
Any signals routed to this modules input slots will be passed on to the host DAW. Note that there’s a built in brick wall limiter at 0dB in this module.
5 - Host parameters
These are the 8 parameters known by the host (that also can be automated). A module called ”Host Parameters” can be added to the surface and each one of those outputs reflects the DAW parameters.
A parameter can be renamed by clicking its name and type a new one. Note that the host won’t see this change, the name is only visible inside the plugin.
6 - Contextual info
When hovering a module with the mouse some more info about the module will show up here.
7 - Window resizing
Click and drag to resize the window.
8 - Annotations
By right-clicking the canvas, one can also create an annotation. Click the annotation text to change it, and click and drag the left side to move it. These annotations are just comments to help out understanding more complex patches.
9 - Menu
Click to reveal a list of functions to load / save / publish patches.
10 - Patch title
A saved patch’s (via the Menu) name
11 - Share
Opens a saved patch in the default browser. This link can be shared and imported via Menu/Import patch URL
New Patch
Creates a new empty patch.
My Patches
Opens a list of saved patches. Click a patch to load it. Click the rightmost X to remove a patch.
Public Patches
Opens a list of all public patches. Click a patch to load it. The list can be filtered by category and optionally only show trusted patches (created or confirmed by Klevgrand) and/or favourited patches. To make a patch a favorite, tick the corresponding box in the ”Fav” column.
Save Patch
Saves a patch. If the patch name already exist in the My Patches list, it’ll be overwritten. If the Public checkbox is set, the patch will be visible to all users in the Public Patches list.
Import Patch URL
Import a patch using a direct URL. This function also imports non-public patches.
Logout
Logs out the current user.
This package contains a set of example patches. In the tutorial folder the following patches can be found.
This is a simple patch that maps a host parameter value to two gain modules via a splitter module. By adjusting the parameter (from host or from the plugin UI) the output gain will change between 0 and 1 (-inf dB and 0 dB).
This module creates a tremolo effect (output gain is controlled by an oscillator). Note that the output of the oscillator is between -1 and 1, so the transform module ”Audio to Fac” is used to scale to a value between 0 and 1.
This value is then multiplied with a user defined ”depth value”, and then inverted making a minimal depth returning a value near 1 instead of 0.
A haas delay takes a stereo signal and delays the left or right channel a few milliseconds to achieve a psychoacoustic effect making it sound panned to the opposite side of the delayed signal.
This example maps two host parameters scales them to a value between 0 and 0.01 (0 and 10 milliseconds) and uses these values as delay time. Use the host parameters ”Delay Left” and ”Delay Right” to control each channels delay time separately.
A feedback delay takes a fraction of its own delayed signal and adds it to the input of the delay.
This example converts the input signal to mono, and then splits it to a dry and a delayed signal. The merger before the delay mixes the regained (using a Multiply module) feedback signal with the to-be-delayed signal.
The wet output is controlled by an additional gain (wet level), and in parallell the dry signal is gained before they’re merged into one signal for output.
This example shows an alternate way of generating a sine wave using math functions and it also uses the Beat Sync module for synchronization with the host tempo and position.
The Beat Sync module outputs the current song position in beats (and fractions of a beat). We only need the fraction so the Modulo module to the right takes care of that.
Then the value gets multiplied with a User Value, which sets number of cycles of the oscillating wave per beat. This value must also be Modulo:ed with 1 to make it possible to convert safe to a radian range (Fac 2 Radian).
We’re using cosine instead of sine because it starts with 1 (sine starts with 0). This makes wave peak at the start of the beat.
The cosine value is then transformed to a factorial value between 0 and 1, and then scaled by a User Value. Before the split, the value is inverted (just as in the previous Tremolo example).
Modules that generates or processes audio in some way.
Frequency in hertz
Phase offset (in radians)
Sync (if >=1, this module will reset)
Generated sine wave (value between -1 and 1)
Audio In
Delay time (in seconds)
Outputs delayed signal
Audio In
Frequency
Q value
Outputs filtered signal.
Audio In
Frequency
Q value
Outputs filtered signal.
Audio In
Frequency
Q value
Outputs filtered signal.
Audio In
Frequency
Q value
Outputs filtered signal.
Audio In
Gain level (factor)
Amplified signal
Gain level (factor)
Noise signal
Data Input (Input signal)
Attack Time (seconds)
Release Time (seconds)
Envelope signal
Audio Input
Threshold (in dB)
Ratio (should be larger than 1)
Knee (in dB)
Gain in dB
Input:
Audio Input
(Audio) Band 1, Band 2, Band 3, Band 4
Audio In
Frequency
Q value
Outputs filtered signal.
Audio Signal
Threshold (linear value, not dB)
Softness (gain map curve)
Processed audio Signal
Audio Signal
Gain Level (linear)
Curve (exponent)
Processed audio signal
Audio Signal 1
Audio Signal 2
Mix Value (0-1 where 0.5 blends both equally, 0 only outputs Sig1 and 1 Sig2)
Blended audio signal
Helper modules for signal routing etc.
Modules outputting values provided by the host application.
changes to these parameter values are smoothed with a slew function.
These modules perform basic math operations and always have two inputs and one output.
A set of common math functions.
These modules only have one parameter output. These values are static and won’t change (except for ”User Value” that can be changed by the user).
0 - Outputs zero.
-1 - Outputs -1
1 - Outputs 1
2 - Outputs 2
PI - Outputs PI (3.141596…..)
sqr(0.5) - Outputs the square root of 0.5
sqr(2) - Outputs the square root of 2
Helper modules. The result of all these can be calculated with ordinary math modules, but these modules can be used to make the canvas more clean and easy to overview.
Converts a factorial (between 0 and 1) to a frequency between 20 and 20000.
Converts a frequency value to a factorial (between 0 and 1).
Converts a factorial value to an audio signal value. Output = 2*Input - 1
Converts an audio value to a factorial value (Output = 0.5*Input + 0.5)
Converts a factorial value to radian range (Output = PI2Input)
Converts a radian value to factorial value (Output = input / (PI*2))
Inverts a factorial value (Output = 1-Input)
These modules outputs values are changing in different ways and can be used for modulation, signaling, smoothing and timing.
Work in progress. Does not currently work.
This category contains modules that handles conditions and triggers.
if (A<B) return C; else return D;
if (A>B) return C; else return D;
Different ways of showing a data signal visually. These modules endpoints must be the output module (or a Null module!) to be able to show any values.
Shows the signal average value as text, and also indicates its value graphically if the value is near audio level. Middle equals 0, left side -1 and right side +1.
Displays a heartbeat-monitor-like curve over the incoming signal. Values (0-1)