Serum 2 User Guide
Serum 2 User Guide
Serum 2
Version 2.0.18
Manual Version: 1.0.3
April 27, 2025
Oli Cash
Nick Dowell
Steve Duda
Dave Gamble
Damon Hancock
Lance Thackeray
Matt Aimonetti
Laurent de Soras
Andrew Simper
Yan Lhert
Jeff Rona
Joel Zimmerman
David Alexander
The information contained in this document is subject to change without notice. In no event shall Xfer
Records or the author of this document be liable for any damages arising out of or related to this docu-
ment or the information contained within it. No part of this document may be reproduced or transmit-
ted in any form or for any purpose without the express written consent of Xfer Records.
WWW.XFERRECORDS.COM
Exploring Serum 16
Exploring Sound Design in Serum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Getting Started 18
Adding Serum to a Track . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Loading a Serum Preset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Creating a New Sound . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Saving Changes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Embedding Content When Saving a Preset . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Dragging Audio to Your DAW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Exploring Basic Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Displaying Help (Tooltips) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Using the Serum Keyboard . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Using Knobs and Sliders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Undo and Redo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Controlling the Main Output Volume . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Using Oscillators and Filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Enabling an Oscillator or Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Choosing Oscillator or Filter Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Using Pitch Controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Routing an Oscillator or Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Accessing the Oscillator or Filter Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Resizing the UI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
Note: Throughout this manual, we’ll refer to Serum 2 (the latest version of Serum) simply as Serum.
Serum combines wavetable, subtractive, multi-sampled, sampled, granular, and spectral synthesis into an
intuitive and cohesive workflow. This manual helps you get the most out of your instrument.
Registering Serum
If you purchased Serum through Xfer Records, you are automatically registered for free updates.
If you purchased Serum through Splice.com, registration is automatic with the email you used at the
time. If you are having trouble locating your registration on xferrecords.com, use the contact form
and mention your Splice username.
Getting in Touch
If you have any feedback about Serum, wish to pass on your comments regarding the software or this
manual, or would like to send links to cool tunes made with Serum, please reach out to:
https://support.xferrecords.com
Downloading Serum
To download Serum, navigate to www.xferrecords.com, sign in to your account, and choose Your
Account in the user menu.
Installing Serum
Serum is available for both Apple macOS and Microsoft Windows. Serum is designed to be used within
a host audio application that supports the VST3, AU, or AAX plug-in formats.
Important: You must select the VST3 version when installing the AU or AAX version of the plugin.
Refer to the documentation that came with your DAW for details.
C:\Users\(username)\AppData\Roaming\Xfer\
If you have Microsoft OneDrive enabled, delete the Serum 2 Presets folder in the following
location:
C:\Users\(username)\OneDrive\Documents\Xfer\
C:\Users\(username)\Documents\Xfer\
Note: Serum does not modify the Windows registry, therefore, you do not need to run an uninstaller
application to completely remove Serum from your system.
• /Library/Audio/Plug-Ins/VST3/
• /Library/Audio/Plug-Ins/Components/
• /Library/Audio/Presets/Xfer Records/
• /Library/Application Support/Avid/Audio/Plug-Ins/
• ~/Library/Preferences/Serum2Prefs.json
• ~/Library/Application Support/com.xfer.serum2/Serum2.lic
Whether you’re designing cutting-edge textures or classic synth tones, Serum offers the flexibility to
explore both familiar and experimental sonic territories with ease.
In addition, Serum features dual filters, each offering a wide selection of filter types for precise tonal
shaping. You can route oscillators through one or both filters, giving you detailed control over the
harmonic structure and timbre of your sound.
Serum Interface
Complementing this, Serum includes an array of fully-configurable envelopes and LFOs (Low Frequency
Oscillators) that you can assign to almost any control. This allows for intricate, evolving soundscapes
with dynamic changes and rhythmic modulations.
To further expand the sonic possibilities, Serum integrates a sophisticated arpeggiator and MIDI clip
system, offering advanced sequencing and performance options for live and studio use.
Taken together, the Serum interface is conveniently structured to help make sound design both easy
and intuitive.
Specifically, Serum offers you the following sound generation options, accessible through oscillators A
to C (OSC A, OSC B, and OSC C), with each offering a distinct approach to sound sculpting:
• Multisample — Uses multiple recordings (samples) of instruments (or other sound generators)
across various notes and dynamic ranges
• Sample — Uses single samples that can be played back at different pitches, layered, looped, or
transformed with various effects
• Granular — Layers, stretches, rearranges, or modulates tiny sound fragments called grains to create
new sound textures
• Spectral — Manipulates sound by analyzing and resynthesizing the frequency spectrum, creating
new sounds with independent control over time and pitch
Serum Oscillators
Similarly, you should hear the sawtooth sound if you click any of the piano keys at the bottom of the
Serum user interface (UI).
Note: If you are not sure how to add an instrument to a track, refer to the documentation that came
with your DAW.
After choosing a preset, you can click the < > arrows (to the right of the preset name) to navigate
through a particular preset subfolder without having to repeatedly display the presets menu.
You can also choose a preset by clicking the button (next to the main menu near the top right) to
access the presets browser.
Click an entry in the list to load the corresponding preset. Most presets load immediately; presets with
larger embedded samples, such as multisampled instruments, display a small green progress bar (directly
beneath the preset name) when loading.
You can preview presets to help you quickly find the right sound.
You can choose the fallback clip from among three standard options. Serum also allows you to auto-
preview clips. See “Performing Standard Preset Operations” on page 334 for more information.
Preset Playing
If you just added Serum to a track, you will see the - Init - preset, which has a single wavetable oscillator
enabled (OSC A) together with basic envelope and LFO modulators defined. This is a perfect blank slate
from which to craft and evolve your sound.
Note: If you’ve already loaded a preset or made other changes and would like to start over, click the
main menu and choose Init Preset in the context menu.
Saving Changes
After crafting a new sound or modifying an existing preset, you can save the configuration as a new
preset (that you can load later).
a. Double-click the ARTIST field directly below the preset name and type the artist name.
b. Double-click the DESC field and add any relevant information related to the preset.
You can add any type of information that you consider helpful or informative.
A dialog appears allowing you to specify the file name and choose the location. Note that this file
name becomes the preset name in Serum.
After you’ve already saved a preset and made modifications, click the (disk) button to save
your changes. A dialog appears allowing you to type a new file name or overwrite your existing
preset file.
If you already know that you want to use the same file name (and overwrite the existing preset
file), Option-click (macOS)/Alt-click (Windows) the disk button. The preset saves using the
same file name without displaying the dialog.
Embed Indicator
Hover over the left of the Serum logo, and drag the wave
icon that appears to an audio track in your DAW.
Option-click (macOS)/Alt-click (Windows) the wave icon to open the Renders folder. Shift-
drag the wave icon to copy the current (saved) preset to the macOS Finder/Windows Explorer.
This section describes common operations when using the on-screen controls, including options
that you can use with most knobs and sliders. Subsequent chapters will describe specific options for
individual controls and other elements of the interface.
Tooltip Help
Serum Keyboard
You can also use the keyboard to specify basic settings including the following:
• TRANSPOSE — The number of semitones (positive or negative) to transpose all incoming and
generated MIDI notes.
• KEY — The key to use throughout Serum, most notably in the CLIP and ARP modules.
• SCALE — The scale to use. Notes outside the scale automatically conform to the selected scale.
• SWING — Use to swing or shuffle the timing of the grid in the CLIP and ARP editors.
• OSC MAPPING — Edit the note and velocity ranges of the oscillators and arpeggiator.
You’ll also see the following options available on nearly all knob and slider menus:
• MIDI Learn — Activates MIDI learn mode. When enabled, Serum waits for an incoming MIDI CC
value.
After Serum receives a MIDI CC value, MIDI learn mode is deactivated and the CC# is assigned to
the knob or slider. Note that the assignment is saved with the preset (patch).
• Lock Parameter — When enabled, locks the control setting (preventing a value change) when
loading presets. You can, however, continue to adjust the control manually.
The Reset Control and Lock Parameter options appear in the context menu of almost every
control in Serum.
In all cases, you can use these options to reset the control to the default value and lock a
control parameter to prevent it from changing when loading presets, respectively.
When saving a DAW session, MIDI CC assignments are saved and recalled with the
session.
When saving a preset, MIDI CC assignments are saved with the preset, but are only loaded
if the Load MIDI Map from Preset preference is enabled on the Global page (this setting is
disabled by default).
See “Preferences” on page 313 for more information about setting global preferences.
You can set the current MIDI CC assignments to load by default by choosing Save MIDI
Map in the main menu and saving the MIDI map as default.SerumMIDIMap in the
Serum 2 Presets/System/MIDI CC Maps folder.
This map then loads automatically when creating a new instance of Serum or choosing Init
Preset in the main menu. The map also automatically loads when selecting a preset if the
Load MIDI Map from Preset preference is disabled.
You can access the undo and redo buttons in the Serum header, near the top right.
Click the button to undo your last operation (such as a change to a knob or an LFO shape,
among others).
Click the button to redo the last undo operation. This allows you to quickly compare and evaluate
changes to your sounds side by side.
Main Volume
Each oscillator type features specific settings and parameters that you’ll read about later in this guide.
This section describes a series of operations that are common across most oscillator and filter types.
You can also use the power button to mute (disable) an oscillator to either solo the other oscillators or
free up CPU, as needed. Notice that when an oscillator is off, the entire panel is dimmed.
Enabling a filter is very similar, except that you need to click the power button directly.
Choosing a Wavetable
Serum uses separate controls for the four settings to facilitate automation and modulation (rather than a
combined value such as 36.04).
It’s often handy to assign an LFO to the octave setting, for instance, without having to count in 12
semitones.
There are times, however, when you might want to have an LFO control an oscillator pitch in a more
coarse manner (a siren-type sound might require an LFO to modulate an oscillator pitch smoothly across
an octave or more).
You can do this by modulating the CRS setting. (You’ll read about modulating controls later in this guide.)
Use the Global > Main Tuning modulation destination to have all oscillators follow a coarse
pitch change.
• Harmonics — Change the pitch by multiplying the base frequency using whole number harmonics.
This generates pitches based on the harmonic series, which is useful for creating overtone-rich
sounds, like organ tones or harmonic layers.
• Ratio — Set the pitch of the oscillator in relation to the base frequency using ratios, which is
common in FM synthesis.
In this case, the oscillator is tuned to a specific ratio relative to another oscillator to create complex
timbres. When you select this mode, you can then set the specific source (SRC) and ratio.
• Step — Adjust the pitch up or down in periods and steps, as defined by the active MTS-ESP tuning.
This option only appears when you are using microtuning and have an MTS-ESP tuning source
available.
Target Description
Filter Route the signal to either FILTER 1, FILTER 2, or by varying degrees to both.
Main Route the signal to the main output, passing through the effects section.
Direct Route the signal to bypass the filter and effects section and play “clean”
along with the main output.
None Route the signal to no output path. You can use this setting when you
want to use an oscillator as a modulation source without having the sound
included as part of the output signal.
You can access the routing options by clicking the button near the top right of an oscillator or filter.
Routing Menu
Oscillator Menu
Locking a Module
You can lock a module by right-clicking the module label and choosing Lock Module in the context
menu (see the Oscillator Menu above). This causes the parameters (settings) of the oscillator module to
remain unchanged (locked) as you change presets.
For example, if you lock OSC A, the module remains the same even when you load a preset (that would
normally change the oscillator module).
After locking a module, you can unlock individual controls (such as the WT POS knob, for
example), as needed.
Note: This does not apply when you initialize the entire preset (by choosing Init Preset using the main
menu). This initializes all settings and removes any module locks that you might have set.
Initializing a Module
After making changes to an oscillator or filter, you can return the module to an initialized state without
affecting any other Serum setting.
Right-click the module label and choose Init Module in the context menu. The oscillator returns to initial
settings (including removing all modulators assigned to the module).
Copying a Module
You can copy an oscillator or filter (with or without modulations) and paste it to another (similar) module.
This works when copying and pasting modules in the same preset, as well as when you want to copy
and paste between presets.
For example, consider the case where you configure an oscillator module with specific wavetable,
unison, detune, and warp settings. In addition, you have LFO 1 modulating the wavetable position
(WT POS knob).
Right-clicking the oscillator label and choosing Copy in the menu copies the module to Serum’s
internal clipboard. Right-clicking another oscillator module and choosing Paste in the menu pastes the
configuration (all settings except the WT POS modulation) to the corresponding module.
To copy and paste a module configuration including modulations, right-click the oscillator label and
choose Copy (w/mods) in the menu. Pasting to a new module now includes the WT POS modulation
from the earlier example.
As mentioned, this copy and paste operation works even after changing or initializing a preset. In
addition, the module configuration stays on Serum’s internal clipboard after pasting, allowing you to
quickly paste the same configuration to multiple modules as needed. It is not possible, however, to copy
a module from one instance of Serum to another.
Hold the Option/Alt key and drag the module label (next to the power button) to another
(similar) module to copy the module without modulations.
Hold the Shift-Option or Shift-Alt keys and drag the module label to another module to copy
with modulations.
Dragging from one module label to another without any keyboard modifiers swaps the two
modules (including modulation assignments).
Pitch tracking instructs the oscillator to adjust its pitch in response to the MIDI note or key being
played. This ensures that the oscillator’s frequency corresponds to the desired musical pitch. Pitch
tracking is typically used with melodic and harmonic sounds, where the pitch needs to follow the
keyboard.
You can choose to disable pitch tracking by right-clicking the oscillator label and toggling Enable Pitch
Tracking off in the context menu.
You might choose to disable pitch tracking with the following types of sounds:
This produces a constant, unchanging pitch regardless of the notes played. This is common with
ambient soundscapes that don’t need pitch variation. This is also useful for layering a static tonal
element beneath a dynamic lead or pad.
• Percussive sounds
These sounds often don’t rely on pitch tracking since their character is defined more by their
transient and timbral qualities than specific pitch. You can use this with a variety of drum sounds,
such as kick drums, snares, or hi-hats, as well as metallic or inharmonic percussive textures.
• Noise-based effects
Noise signals (white, pink, or custom noise) aren’t inherently pitched, so pitch tracking is irrelevant.
You can also use this with wind, rain, or static effects, in addition to risers, sweeps, and impacts.
• Experimental sounds
Disabling pitch tracking can lead to unexpected and unique sonic results. This can be helpful for
creating unconventional or dissonant sounds. This can more easily allow you to explore textures
where the focus is on timbre and modulation rather than pitch accuracy.
You can also disable pitch tracking to add layers without harmonic conflicts. A non-pitch-tracked
oscillator can add texture or depth without interfering with the harmonic structure (such as a static sub-
bass tone underneath a harmonic element).
When pitch tracking is disabled, the Multisample, Sample, Granular, and Spectral
oscillators play C3 (MIDI note 60), whereas the Wavetable oscillator plays C-2
(MIDI note 0), allowing it to be used as an LFO.
Resizing the UI
You can resize the Serum user interface to make it fit appropriately
within your work environment.
The easiest way to resize the interface is to click and drag the lower
right corner of the UI (similar to how you would resize most other
windows using your computer).
For more precise control, click the Serum 2 logo (near the top left)
and choose a resize option in the menu that appears. The window
resizes to match your selection.
After resizing, you can make this size the default setting by clicking
the Serum 2 logo again and choosing Set x% as Default, where x is
the current setting.
To return to the default setting of 100%, click the Serum 2 logo and
choose Default (100%) using the menu.
Resize Menu
Serum oscillator playback has been carefully constructed to give you a high-frequency representation to
the limits of the human ear, without the audible aliasing artifacts (Nyquist reflection) commonly found
on most wavetable synthesizers.
While this requires more CPU during both load and run-time, Serum’s advanced SSE (Streaming SIMD
Extensions) optimizations effectively help to minimize the CPU expense. We think you’ll agree that the
benefits in sound purity make it worthwhile.
What is a Wavetable?
A wavetable is a small amount of digital audio (sample data or waveform) that is played back in a looping
fashion.
The frequency (pitch) of the resulting note is determined by the rate (the speed between the repeats)
at which the waveform is played. The tone (timbre and harmonics) of the sound is determined by the
content of the waveform.
Strictly speaking, when Serum loads a wavetable, it uses 2048 samples for a frame
(subtable) of the wavetable set. This means that the maximum file size is 2048 (samples) x
256 (frames) x 32 (bits), which is exactly 2 megabytes.
However, most wavetable files will not be this large. A good sounding wavetable can consist
of just a few frames. The remaining frames can be interpolated (in the Wavetable Editor) to
allow for smooth-sounding transitions.
These interpolated frames are generated through crossfading (mix blend) or spectral
morphing (frequency + phase blend). These frames are computed at load time; Serum
embeds the interpolation type rather than the interpolated waveforms (reducing disk
space).
Using Wavetables
Serum features three multi-purpose oscillator
modules, as well as dedicated NOISE and SUB
oscillators (discussed separately later).
OSC A in 2D View
Choosing a Wavetable
Editing a Waveform
You can take a closer look at a wavetable by hovering over the top-right corner of the waveform display
and clicking the button.
Wavetable Editor
Phase
Use the PHASE control to specify where the
oscillator begins playing back when a note is
triggered.
Having said that, the effect can be very pronounced in the following scenarios:
If your wavetable is a sine wave, for instance, and the ENV 1 attack is very fast (set to a small
value), you will hear a click at the beginning of the sound if the PHASE control is set to a non-zero
crossing (0% or 50% represent the zero crossings in a default sine wave).
The interaction between two or more oscillators can be very noticeable due to phase cancellation.
Adjusting the start time of one of the oscillators results in a different tone.
If the oscillator waveform display is in 2D (single table view), moving the PHASE control causes a
yellow line to appear indicating where the note onset will occur.
For example, do the following to hear the effects of the RAND control:
At this point, you should hear a “laser zap” phasing sound resulting from the unison oscillators all
starting together in phase, and slowly drifting apart from their detune.
Every time you trigger the note, you’ll hear the same laser zap sound since the voice phases are
restarting. While this can sometimes be cool, generally this is undesirable as the sweeping sound can be
distracting.
As you raise the RAND control, notice how the effect becomes less pronounced as a random phase
offset is introduced into each voice separately. By the time you reach 100% using the RAND control,
you will no longer hear the “zap” sound.
Phase Legato
In the case of the “laser zap” phasing sound produced in the previous section, the default behavior is
that when Legato is disabled (in the VOICING section) and a voice is stolen, the wavetable oscillator
phase does not reset at note on. Therefore, you won’t hear that “laser zap”.
Also, in cases when Legato is enabled and Serum is playing polyphonically (including when not stealing
a voice), the wavetable oscillator phase does reset at note on. This results in you hearing the zap even
though legato is enabled and you want a smooth transition between notes.
The default behavior is how Serum 1 works, and is intended to avoid clicks at note on, which would be
particularly evident with a simple sine wave.
However, with the above settings, we want the opposite behavior. You can toggle this by right-clicking
the PHASE control and choose Phase Legato in the context menu.
Option Description
All Voices New notes use the PHASE and RAND control settings for all voices. This is
the default setting.
Contiguous New notes continue with the phase of the previous note.
Per Voice New notes start with the same editable phase each time.
If you set phase memory to Per Voice while one or more notes are sounding, Serum captures
the current phase from the most-recently sounded note.
Wavetable Position
Use the WT POS knob to set the position within
the wavetable. In other words, the knob selects
the frame (subtable) that is currently audible.
Setting Description
This mode can sound very controlled and smooth, giving a thick and
coherent texture.
• Super — Multiple voices are slightly detuned from each other but with
a special emphasis on creating a dense and powerful sound, often with
a slight stereo spread.
Use this mode to create a lush, wide sound, especially with supersaw
sounds, where the detuned voices give a rich, full sonic character.
This mode can create unique, rich textures, especially for sounds that
need to be more aggressive or have a strong presence in a mix.
Setting Description
MODE (cont.) • Inv — The detuning behavior is inverted, with lower voices detuned
more sharply compared to higher ones.
This can create a more organic or chaotic sound, ideal for achieving
textures that are less polished and more natural or experimental.
STACK The unison stacking, which instructs Serum to not just duplicate the unison
voices at the same pitch but instead transpose or harmonically adjust the
voices.
• 12+7 (1-3x) — Distribute voices between the original pitch, fifth, and
octave transpositions, with a range 1-3 octaves higher
When you increase the width parameter, the unison voices are spread
farther apart in the stereo spectrum, placing some voices more to the left
and others more to the right. This creates a sense of space and a wider,
more immersive sound.
Conversely, decreasing the width narrows the stereo spread, making the
sound more centered and focused. This can be useful for sounds that need
to be more direct or fit better in a dense mix without dominating the stereo
field.
Setting Description
RANGE The extent or range of detuning applied to the unison voices, determining
how far apart in pitch the individual voices are spread around the central
frequency of the original sound.
When the range is set to a low value, the detuning between the unison
voices is minimal. The voices remain relatively close in pitch to the original
frequency, creating a subtle and smooth chorusing effect. This can add
warmth and a slight thickness to the sound without drastically changing its
character.
You can further adjust the unison setting using the DETUNE and BLEND knobs.
Unison Detune
Use the DETUNE knob to specify the tuning offset +/- for the additional voices. This is only applicable
when unison is enabled (set to a value above 1).
Unison Blend
Use the BLEND knob to specify the level offset of the unison voices versus the “central” unison voice or
voices (1 if unison is set to an odd number of voices and 2 if set to an even number).
You can think of BLEND as a wet/dry mix between a unison (wet) and non-unison (dry) sound. The
default value of 75% is an even blend between all the voices. Note that this is only applicable when the
number of unison voices is greater than two.
Warp
Setting the warp allows you to manipulate the
playback (and sound) of the wavetable oscillator.
Alt Warp Mirror Create a mirror-image of the waveform for the second half of
(cont.) the duty cycle.
This has an “octaved” type of quality to the sound due to the
doubling of the waveform into both halves of the wave cycle.
For this setting, the WARP knob behaves similarly to the Asym
+/- mode, except on both halves independently.
Due to the mirroring of the waveform, this mode always has
an audible effect.
Remap 1 Custom remapping of the wave cycle. When you select this
option, a pencil button appears that you can use to open a
graph showing the way your waveform will remap.
A diagonal line from bottom-left to top-right indicates no
change to the waveform (y=x). The WARP knob determines
the strength of the remap, from 0 (y=x) to 100% (what you see
on the graph).
Remap 2 Represents mirrored remapping.
This is the same as Remap 1 but applies the graph to each
half the waveform independently. This allows for symmetric
remapping without the need to draw symmetric shapes on the
graph.
Remap 3 Represents sinusoidal remapping.
This is another remapping option that saves you from having
to draw fancy curves.
Remap 4 Represents a 4x remapping.
This is similar to Remap 2 (mirrored) but in this case the graph
applies four times. This creates a more busy sound, and can be
helpful when you want something nasty.
Quantize Similar to sample-and-hold, that is, a sample rate reduction.
Compared to an SR Redux effect, this applies to the waveform
itself.
This causes the aliasing sound to follow the pitch perfectly
(instead of having that “same ringing pitch on all notes” quality
that a redux effect creates).
Odd/Even Proportionally vertically scale the waveform. Use this to emit
only the odd or even harmonics from the signal or a mixture of
the two.
At 50%, you have the original signal. At 0%, you hear only the
odd harmonics. At 100%, you hear only the even harmonics
(which creates an octaving effect since the first harmonic is
missing).
Distortion Zero-Square Modify the waveform in such a way that any part of the signal
(cont.) below a certain amplitude threshold is forced to zero, while
parts of the waveform above the threshold are squared or
otherwise drastically altered.
This creates a sharp, abrupt change in the shape of the
waveform, leading to a sound that is both harsh and
harmonically rich.
Asym Create an asymmetric effect that applies different distortion
characteristics to the positive and negative halves of an audio
waveform.
This type of waveshaping introduces a unique harmonic profile
by treating one side of the waveform differently from the
other, resulting in a sound that can range from subtle warmth
to highly complex and rich overtones.
Rectify Create a type of waveshaping effect that modifies the signal
by altering or “rectifying” the waveform, typically by flipping or
removing one half of the waveform.
This creates a distinct, harmonically rich, and sometimes harsh
or metallic sound, often associated with aggressive or synthetic
timbres.
Sine Shaper Create a type of waveshaping distortion that shapes the audio
signal using a sine function.
This type of distortion applies a nonlinear transformation
to the input signal, resulting in a smoother, more rounded
distortion that introduces harmonics in a musical and often
warm manner.
Stomp Box Create a distortion effect that emulates the sound of classic
distortion or overdrive pedals used by guitarists (stomp boxes).
This effect reproduces the gritty, crunchy, and saturated tones
characteristic of analog guitar pedals, bringing warmth, edge,
and intensity to a wide range of sounds.
Tape Sat. Create a distortion effect that emulates the warm, rich sound
characteristics of analog tape recording.
Tape saturation is a form of soft-clipping distortion that occurs
naturally when audio signals are recorded to magnetic tape,
especially at higher levels.
This effect is highly valued for its ability to add warmth,
harmonic richness, and a sense of vintage character to an
audio signal.
Distortion Soft Sat. Create a distortion effect that introduces subtle, smooth,
(cont.) and musical saturation to the signal. It is designed to gently
enhance the harmonic content of the sound without
introducing harsh or aggressive distortion.
Soft saturation is often used to add warmth, fullness, and a
natural, analog-like character to audio, making it a popular
choice for enhancing digital recordings.
FM FM (from other Perform frequency modulation using the other oscillator or
oscillator) filter.
For example, FM The other oscillator or filter must be enabled for this to work,
(B) however, you can turn down the volume of the other oscillator
if you want to use the other oscillator simply as a modulation
FM (from other
source.
oscillator)
For example, FM
(C)
FM (Noise)
FM (Sub)
FM (Filter 1)
FM (Filter 2)
Thru-Zero Cause the carrier oscillator (the one producing the sound)
to continue to oscillate correctly even when its frequency is
modulated into negative values by the modulator oscillator.
Normally, if the modulation drives the carrier oscillator’s
frequency below zero, it either clamps at zero (stops
oscillating) or reflects back to a positive value (causing a
discontinuity).
With Thru-Zero, when the frequency modulation drives the
carrier frequency below zero, the carrier oscillator doesn’t stop
or reflect; instead, it inverts its phase and continues oscillating.
This allows for smooth and continuous modulation, resulting in
a more natural and harmonically rich sound.
The inversion of phase (caused by negative frequencies) adds
new harmonic characteristics, making Thru-Zero especially
useful for lush, metallic, or bell-like tones.
FM (cont.) Exp Use an exponential scaling curve. This means small changes
in the modulator’s amplitude can cause dramatic changes
in the carrier frequency, especially as the modulation depth
increases.
Compared to linear FM, exponential FM produces a broader
and more pronounced harmonic spectrum. It is often described
as brighter or harsher due to the rapid frequency sweeps
caused by the exponential relationship.
Linear Use a linear scaling curve. This means that the modulation
source (modulator oscillator) affects the frequency of the
carrier oscillator in a direct, proportional manner.
This ensures that the carrier oscillator maintains its overall
pitch, even when modulated heavily. This makes it particularly
useful in musical contexts, as it allows for predictable harmonic
and inharmonic spectra without drastically detuning the base
pitch.
Linear is often described as smooth, clean, and more “musical”
compared to other FM types such as exponential FM. This
makes it great for bell-like tones, pads, and other complex but
stable timbres.
Note that linear FM is thru-zero but with a clamp at zero. This
allows you to do the traditional “can’t do thru-zero” FM.
PD PD (from other Perform phase distortion using the other oscillator or filter.
oscillator)
The other oscillator or filter must be enabled for this to work,
For example, PD however, you can turn down the volume of the other oscillator
(B) if you want to use the other oscillator simply as a modulation
source.
PD (from other
oscillator) This is similar to FM except that the phase is modulated
instead of the frequency.
For example, PD
(C)
PD (Noise)
PD (Sub)
PD (Filter 1)
PD (Filter 2)
PD (Self)
Pan
Use the PAN knob to control the placement of the
waveform in the stereo field (left to right).
Level
Use the LEVEL knob to control the output volume
of the oscillator.
With a multisample instrument, individual notes are recorded at different velocities (soft to loud) and
sometimes with various playing techniques, such as plucking, bowing, or using different mallets.
When using a multisample instrument, Serum accurately selects the appropriate sample based on the
MIDI input—evaluating both the pitch and velocity of the note—ensuring that the output sound mimics
the real instrument’s response as closely as possible. For example, playing softly might trigger a sample
of a gently played note, while pressing hard would trigger a sample of the same note played with force.
Layering multiple distinct samples achieves the effect of rich, lifelike audio textures that respond
accurately to different playing styles. This can capture the nuanced tonal variations and expressive
characteristics of the original instrument, allowing for a highly realistic and responsive musical
experience.
Multisample Instrument
Knob Description
DELAY The time (in milliseconds) before the envelope begins after a note is
triggered, allowing you to introduce a pause before the attack phase starts.
This is useful for creating rhythmic effects or gradual sound layering, giving
more control over how and when the envelope affects the sound.
Right-click the DELAY knob and choose BPM Sync in the context menu to
synchronize the delay with the musical tempo. When selected, you can set
the delay in beats and bars.
A (Attack) The time it takes for the sound to reach its maximum level after a note is
triggered. This shapes the initial onset of the sound, allowing for smooth
fades or sharp, immediate beginnings depending on the setting.
H (Hold) The amount of time the sound remains at its peak level after the attack
phase is completed.
This allows the sound to sustain momentarily before moving into the decay
phase, adding emphasis and length to the peak of the sound.
D (Decay) The time it takes for the sound to transition from its maximum level after the
attack phase to the sustain level.
This shapes the gradual reduction in volume, allowing you to create more
natural fades or sharp drops in sound intensity depending on the setting.
Knob Description
S (Sustain) The steady level the sound maintains after the decay phase, as long as the
note is held.
Unlike other stages, the sustain level lasts indefinitely until the note
is released, allowing for either continuous sound or a softer presence
depending on the setting.
R (Release) The time it takes for the sound to fade out after the note is released.
This shapes the tail of the sound, allowing for smooth, gradual decays or
quick cutoffs depending on the chosen release time.
You can also adjust the envelope by clicking and dragging each point directly on the graph. The cursor
changes to indicate whether the point can be moved horizontally or freely in all directions, providing
visual feedback for precise control.
Velocity Track
Use the VEL TRACK control to adjust the sensitivity of the note velocity. After enabling velocity
tracking, click and drag in the field to modify the setting.
Random
Use the RAND control to randomize the initial phase of the sample. This is similar to the random setting
for other oscillator modes. Click and drag in the field to modify the value.
You can also adjust the pan and level of the signal.
Multisample Parameters
Timbre
Use the TIMBRE knob to adjust the multisample timbre.
For multisamples with zone mapping, this control inversely adjusts mapped samples to pitch. This alters
how the samples respond to changes in pitch across different key zones.
Normally, as you play higher notes, the pitch of the sample increases, and as you play lower notes, it
decreases. When you inversely adjust the mapped samples to pitch, the opposite happens: higher notes
trigger samples that are mapped to lower pitches, and lower notes trigger samples mapped to higher
pitches.
This creates an unusual or experimental timbral effect, as the relationship between pitch and sample
playback is reversed, potentially adding unique tonal qualities to the sound.
Unison
Use the UNISON control to set the number of unison voices, effectively “stacking” oscillators in a way
that is similar to playing multiple notes of the same pitch, but slightly detuned.
Click the field and drag to set the appropriate value. You can also double-click the field and type a value.
Setting Description
This mode can sound very controlled and smooth, giving a thick and
coherent texture.
• Super — Multiple voices are slightly detuned from each other but with a
special emphasis on creating a dense and powerful sound, often with a
slight stereo spread.
Use this mode to create a lush, wide sound, especially with supersaw
sounds, where the detuned voices give a rich, full sonic character.
• Exp — The detuning between voices increases exponentially as
you move away from the central pitch. This means that the spacing
between voices gets wider more quickly as you move outward, creating
a more dramatic spread compared to linear.
This mode can create unique, rich textures, especially for sounds that
need to be more aggressive or have a strong presence in a mix.
• Inv — The detuning behavior is inverted, with lower voices detuned
more sharply compared to higher ones.
This can create a more organic or chaotic sound, ideal for achieving
textures that are less polished and more natural or experimental.
STACK Set the unison stacking.
Setting Description
WIDTH The extent to which the unison voices are spread out across the stereo field,
determining how wide or narrow the resulting sound feels in a stereo mix.
When you increase the width parameter, the unison voices are spread
farther apart in the stereo spectrum, placing some voices more to the left
and others more to the right. This creates a sense of space and a wider,
more immersive sound.
Conversely, decreasing the width narrows the stereo spread, making the
sound more centered and focused. This can be useful for sounds that need
to be more direct or fit better in a dense mix without dominating the stereo
field.
RANGE The extent or range of detuning applied to the unison voices, determining
how far apart in pitch the individual voices are spread around the central
frequency of the original sound.
When the range is set to a low value, the detuning between the unison
voices is minimal. The voices remain relatively close in pitch to the original
frequency, creating a subtle and smooth chorusing effect. This can add
warmth and a slight thickness to the sound without drastically changing its
character.
You can further adjust the unison setting using the DETUNE and BLEND knobs.
Unison Detune
Use the DETUNE knob to specify the tuning offset +/- for the additional voices. This is only applicable
when unison is enabled (set to a value above 1).
Unison Blend
Use the BLEND knob to specify the level offset of the unison voices versus the “central” unison voice or
voices (1 if an odd number, 2 if an even number of unison).
You can think of BLEND as a wet/dry mix between a unison (wet) and non-unison (dry) sound. The
default value of 75% is an even blend between all the voices. Note that this is only applicable when the
number of unison voices is greater than two.
Warp
Setting the warp allows you to manipulate the playback/sound of the wavetable oscillator.
By default, warp is set to OFF (as displayed next to the corresponding knob). Clicking the current setting
displays a menu from which you can choose from among the available warp modes.
You can also use the < > arrows to conveniently switch between different warp modes without having
to open the menu. After selecting a mode, you can use the knob to set the depth.
See “Exploring the Warp Modes” on page 49 for detailed information about the available warp
modes.
Pan
Use the PAN knob to control the placement of the waveform in the stereo field (left to right).
Level
Use the LEVEL knob to control the output volume of the oscillator.
In addition to the standard features you would expect in a sampler, Serum includes powerful slicing
tools that are integrated with the CLIP mode within Serum, giving you even greater flexibility to
rearrange and reimagine almost any type of audio source.
You can also quickly and easily convert samples to wavetables, offering even more creative possibilities.
When loading a sample, Serum assumes all samples to be tuned to C3 by default (following
the standard that MIDI Note 69 is A3 at 440 Hz).
You can instruct Serum to assume a different tuning by adding the note name at the end of
the file name. For example, naming a sample file Morning Bass F2.flac tells Serum to
set F2 as the root of your sample, removing the need for you to manually adjust the pitch.
By informing Serum about the root note, Serum can now correctly map the sample. When
you trigger F2, the sample will now play at its original pitch, while playing other notes (such
as G2 or E2) will result in Serum pitch-shifting the sample accordingly.
Similarly, you can include flats and sharps in the file name. This means that naming a sample
file Ashes Piano Eb3.flac or Ashes Piano D#3.flac produces the same result; the
choice is yours whether to use flats or sharps.
Important: The note name (appearing at the end of the file name) must be preceded by a
space or underscore character. No other separators are supported.
Serum additionally refers to pitch data embedded in the instrument chunk of a WAV file.
Sample Instrument
Sample Menu
Operation Description
Show Marker Enable to have the start, end, loop start and loop end markers animate to
Animation show the effect of any assigned modulation. Note that dragging the markers
is not allowed when this option is enabled.
Zoom to Start and Reset the display zoom setting to the default showing both the start and end
End markers.
By default, when this setting is disabled (no check mark present), you can
click-drag up and down in the waveform to zoom in and out, and drag left
and right to pan the waveform left and right.
When enabled (showing a check mark), the waveform maintains a consistent
display showing the entire waveform from start to end, without the ability to
zoom or pan using your mouse.
Important: Selecting this option toggles the setting; you need to deselect
this option to allow you to zoom the display again.
Snap Off The next four menu items relate to how the sample playback start and end
points or loop points are adjusted.
With Snap Off, no snapping is applied. You can place the start and end, or
loop points freely along the waveform, without any restrictions or alignment
to specific reference points.
This provides full flexibility when working with waveforms but requires
careful manual placement to avoid clicks, pops, or timing issues.
Snap to Zero Start, end, or loop points snap to the nearest zero-crossing in the waveform.
A zero-crossing is a point where the waveform amplitude is zero.
This prevents audio clicks or pops when the playback starts, ends, or loops,
as abrupt transitions between non-zero amplitudes can create artifacts. This
is ideal for ensuring smooth playback and transitions in the waveform.
Snap to Beats Start, end, or loop points align to the nearest beat grid, based on the tempo
of your track.
This ensures that the waveform points or loops are musically synchronized
with the track tempo. This is useful for rhythmic or tempo-synced loops,
where precise timing is essential.
Operation Description
Snap to Loop The start or end points snap to the nearest pre-defined loop points in the
waveform.
This keeps the points aligned with the loop structure, ensuring seamless
looping without unintended offsets. This is helpful when working on a pre-
looped sample or creating a loop that must align perfectly.
Fade Edges Apply a fade-in and fade-out at the start and end of the sample. You can
choose a setting from 1ms to 128ms as well as None.
This helps smooth out any abrupt changes in amplitude that could cause
unwanted artifacts such as clicks or pops when the sample is triggered or
looped.
Normalize Normalize the sample by adjusting its overall volume to maximize the peak
loudness without introducing distortion. This ensures consistent volume
levels while preserving the original dynamics of the sample.
Reverse Reverse the audio sample.
Trim Trim the sample to the current start and end markers.
Slicing Off
Slice Auto See “Slicing Samples” below.
Slice Manual
The Fade Edges, Normalize, Reverse, and Trim operations are non-destructive to the original
sample file.
You can therefore easily undo and redo the operations using the and button
buttons respectively.
In addition, you can choose Reload Sample in the sample menu if you need to clear all
operations and return to the original sample.
Slicing Samples
Serum can help you slice audio samples into smaller segments, making it easier to trigger specific parts
individually. You can use this technique to create new rhythmic patterns, isolate key elements, or remix a
sample to fit a new sonic context.
• Slice Auto — Automatically slice the sample using a user-configurable threshold (sensitivity)
• Slice Manual — Automatically slice the sample, and then allow you to manually adjust the slices
The process for slicing samples is identical in Sample, Granular, and Spectral modes.
Auto Slicing
Auto Slicing
When you are zoomed in, you can drag left and
right to pan the display.
Manual Slicing
Slicing Options
For example, if you select C3 as the root note, choosing this option assigns the slices to successive
notes in the currently-selected clip, starting at C3.
• Auto-Sync to Clip — Similar to the previous option in that the slices are sent to the currently-
selected clip, with each slice assigned to the corresponding trigger note. However, when you
modify the slices (by changing the slice threshold), the clip is automatically modified to reflect the
new slices.
Loop Menu
Use the loop menu to specify the loop mode for
the sample.
Option Description
One-shot The sample plays forward for the duration of the note.
This technique is commonly used for sounds like drums, percussion hits, or
sound effects, where the full duration of the sample is essential.
Fwd Loop The sample plays from the start marker to the loop end marker, and then
loops back to the loop start marker. This allows you to play the onset of the
audio and then stay sustained in the loop.
This method is ideal for sustaining sounds, such as a held violin note or a
drone, where you want the sample to maintain a consistent, ongoing tone.
Hover over the sample display and set the loop start and end by dragging
the corresponding blue markers. Click and drag to move the loop to a new
location.
Option Description
Rev Loop The sample plays from the start marker to the loop end marker and then
reverses playback direction to loop back to loop start, which then loops
backwards to the loop end.
This creates a unique effect where the sound appears to play in reverse
repeatedly, which can add an interesting, unconventional texture to the
music.
Hover over the sample display and set the loop start and end by dragging
the corresponding blue markers. Click and drag to move the loop to a new
location.
Fwd/Rwd Loop The sample plays in a forward/reverse loop for the duration of the note.
Note that the sample plays from the start marker uninterrupted until you
reach the loop.
This creates a seamless, ping-pong-like effect where the sound alternates
between playing forward and in reverse, providing a smooth and
uninterrupted looping experience.
This type of playback is especially useful for creating evolving and dynamic
textures, as it helps avoid the abrupt transitions or potential clicks that can
occur with traditional forward-only looping.
Hover over the sample display and set the loop start and end by dragging
the corresponding blue markers. Click and drag to move the loop to a new
location.
Tailed The sample plays forward from halfway through the sample to the end (the
tail) and then loops the tail of the sample as the amplitude decays.
This allows the sound to fade out naturally rather than cutting off abruptly.
Relative Loop The looped section of the sample changes dynamically based on the
playback start position.
Rather than always looping between fixed start and end points, the loop
moves relative to the playback start marker.
This can be useful when automating or modulating the start position.
Link Loop Length The loop end marker moves relative to the loop start marker, keeping the
loop length consistent.
This can be useful when automating or modulating the loop start position.
Exit Loop on Release When a key is released and the amplitude envelope is in the release phase,
playback exits the loop and plays to the end of the sample.
It is not possible to drag the loop markers outside the playback start/end markers.
You can, however, drag, automate, or modulate the markers so that the loop end marker is
before the loop start marker. In this case, the loop direction is reversed.
You can also adjust the pan and level of the signal.
Sample Parameters
Scan
Use the SCAN knob to set the speed and direction
of the sample playback.
Unison
Use the UNISON control to set the number of unison voices, effectively “stacking” oscillators in a way
that is similar to playing multiple notes of the same pitch, but slightly detuned.
Click the field and drag to set the appropriate value. You can also double-click the field and type a value.
Note: Unison causes Serum to generate multiple voices, raising CPU usage. The color of the UNISON
field changes as you increase the number of unison voices as a reminder of the CPU consumption.
Setting Description
This mode can sound very controlled and smooth, giving a thick and
coherent texture.
• Super — Multiple voices are slightly detuned from each other but with a
special emphasis on creating a dense and powerful sound, often with a
slight stereo spread.
Use this mode to create a lush, wide sound, especially with supersaw
sounds, where the detuned voices give a rich, full sonic character.
Setting Description
This mode can create unique, rich textures, especially for sounds that
need to be more aggressive or have a strong presence in a mix.
• Inv — The detuning behavior is inverted, with lower voices detuned
more sharply compared to higher ones.
This can create a more organic or chaotic sound, ideal for achieving
textures that are less polished and more natural or experimental.
STACK Set the unison stacking.
WIDTH The extent to which the unison voices are spread out across the stereo field,
determining how wide or narrow the resulting sound feels in a stereo mix.
When you increase the width parameter, the unison voices are spread
farther apart in the stereo spectrum, placing some voices more to the left
and others more to the right. This creates a sense of space and a wider,
more immersive sound.
Conversely, decreasing the width narrows the stereo spread, making the
sound more centered and focused. This can be useful for sounds that need
to be more direct or fit better in a dense mix without dominating the stereo
field.
RANGE The extent or range of detuning applied to the unison voices, determining
how far apart in pitch the individual voices are spread around the central
frequency of the original sound.
When the range is set to a low value, the detuning between the unison
voices is minimal. The voices remain relatively close in pitch to the original
frequency, creating a subtle and smooth chorusing effect. This can add
warmth and a slight thickness to the sound without drastically changing its
character.
Setting Description
START Apply a random offset to the starting position for each unison voice.
This creates subtle timing differences, which can make the sound more lively,
complex, and textured.
SPAN Apply a fixed offset to the starting position for each unison voice.
WARP 1 Spread out the warp amount applied to each voice around the current
WARP 1 knob position.
WARP 2 Spread out the warp amount applied to each voice around the current
WARP 2 knob position.
You can further adjust the unison setting using the DETUNE and BLEND knobs.
Unison Detune
Use the DETUNE knob to specify the tuning offset +/- for the additional voices. This is only applicable
when unison is enabled (set to a value above 1).
Unison Blend
Use the BLEND knob to specify the level offset of the unison voices versus the “central” unison voice or
voices (1 if an odd number, 2 if an even number of unison).
You can think of BLEND as a wet/dry mix between a unison (wet) and non-unison (dry) sound. The
default value of 75% is an even blend between all the voices. Note that this is only applicable when the
number of unison voices is greater than two.
Warp
Setting the warp allows you to manipulate the playback/sound of the wavetable oscillator.
By default, warp is set to OFF (as displayed next to the corresponding knob). Clicking the current setting
displays a menu from which you can choose from among the available warp modes.
You can also use the < > arrows to conveniently switch between different warp modes without having
to open the menu. After selecting a mode, you can use the knob to set the depth.
See “Exploring the Warp Modes” on page 49 for detailed information about the available warp
modes.
Pan
Use the PAN knob to control the placement of the waveform in the stereo field (left to right).
Level
Use the LEVEL knob to control the output volume of the oscillator.
By layering, overlapping, and modifying these grains, you can create complex and evolving soundscapes,
offering a high degree of flexibility and experimentation beyond traditional synthesis methods.
One of the key strengths of granular synthesis is its ability to transform audio in real-time, whether
stretching sounds, altering pitch without affecting duration, or creating rich, atmospheric textures from
even the simplest of recordings. Using the granular mode, you can create everything from shimmering,
ethereal pads to glitchy, fragmented effects, offering limitless possibilities for sonic exploration.
Note: Granular synthesis can be CPU intensive, especially when compared to Wavetable, Sample, or
Multisample modes.
Granular Synthesis
Granular Menu
Loop Menu
Use the loop menu to specify the loop mode for
the sample.
It is not possible to drag the loop markers outside the playback start/end markers.
You can, however, drag, automate, or modulate the markers so that the loop end marker is
before the loop start marker. In this case, the loop direction is reversed.
Window Amount
Click the button to access the grain amplitude
window settings.
Setting Description
AMOUNT Set the influence of the window curve. You can also set the per-grain
randomization for the amount.
Click the corresponding field and drag to set the appropriate value. You can
also double-click the field and type a value.
SKEW Set the window skew. You can also set the per-grain randomization for the
skew.
Click the corresponding field and drag to set the appropriate value. You can
also double-click the field and type a value.
SHAPE Set the grain window shape. A representation of the corresponding shape
appears on the left.
The shape of each grain determines how its volume evolves over time, affecting the attack and decay
characteristics of the sound. By selecting different grain shapes, such as smooth fades or abrupt cuts,
you can dramatically alter the articulation and texture of the sound.
Shape Description
Hann Apply a smooth, symmetrical fade-in and fade-out to each grain, using a
cosine-shaped envelope to taper the volume.
This creates a soft, natural sound with gradual attacks and decays, reducing
any harsh transitions between grains for a more cohesive texture.
Shape Description
Shape Description
Exp Dec Rev Shape each grain with a reversed exponential decay, where the grain starts
at a low volume and rapidly rises to full volume in a curved, nonlinear
fashion.
This creates a grain with a smooth, swelling attack followed by a sharp peak,
ideal for building tension or creating atmospheric effects with a gradual
onset.
Unison
Use the UNISON control to set the number of unison voices, effectively “stacking” granular oscillators in
a way that is similar to playing multiple notes of the same pitch, but slightly detuned.
Click the field and drag to set the appropriate value. You can also double-click the field and type a value.
Note: Unison causes Serum to generate multiple voices, raising CPU usage. The color of the UNISON
field changes as you increase the number of unison voices as a reminder of the CPU consumption.
Setting Description
This mode can sound very controlled and smooth, giving a thick and
coherent texture.
• Super — Multiple voices are slightly detuned from each other but with a
special emphasis on creating a dense and powerful sound, often with a
slight stereo spread.
Use this mode to create a lush, wide sound, especially with supersaw
sounds, where the detuned voices give a rich, full sonic character.
• Exp — The detuning between voices increases exponentially as
you move away from the central pitch. This means that the spacing
between voices gets wider more quickly as you move outward, creating
a more dramatic spread compared to linear.
This mode can create unique, rich textures, especially for sounds that
need to be more aggressive or have a strong presence in a mix.
• Inv — The detuning behavior is inverted, with lower voices detuned
more sharply compared to higher ones.
This can create a more organic or chaotic sound, ideal for achieving
textures that are less polished and more natural or experimental.
STACK Set the unison stacking.
DETUNE Specify the tuning offset for the additional voices.
BLEND Specify the level offset of the unison voices versus the “central” unison voice
or voices (1 if an odd number, 2 if an even number of unison).
You can think of BLEND as a wet/dry mix between a unison (wet) and non-
unison (dry) sound. The default value of 75% is an even blend between all
the voices.
Note that this is only applicable when the number of unison voices is greater
than two.
Setting Description
WIDTH The extent to which the unison voices are spread out across the stereo field,
determining how wide or narrow the resulting sound feels in a stereo mix.
When you increase the width parameter, the unison voices are spread
farther apart in the stereo spectrum, placing some voices more to the left
and others more to the right. This creates a sense of space and a wider,
more immersive sound.
Conversely, decreasing the width narrows the stereo spread, making the
sound more centered and focused. This can be useful for sounds that need
to be more direct or fit better in a dense mix without dominating the stereo
field.
RANGE The extent or range of detuning applied to the unison voices, determining
how far apart in pitch the individual voices are spread around the central
frequency of the original sound.
When the range is set to a low value, the detuning between the unison
voices is minimal. The voices remain relatively close in pitch to the original
frequency, creating a subtle and smooth chorusing effect. This can add
warmth and a slight thickness to the sound without drastically changing its
character.
When the range is increased, the detuning becomes more pronounced,
and the unison voices are spread further apart in pitch. This creates a more
dramatic and sometimes more chaotic or dissonant effect, which can be
useful for creating lush, wide textures or aggressive, detuned leads.
START Apply a random offset to the starting position for each unison voice.
This creates subtle timing differences, which can make the sound more lively,
complex, and textured.
SPAN Apply a fixed offset to the starting position for each unison voice.
SPAWN PATTERN The timing offset at which unison grain voices are spawned.
With Together (default), all unison grain voices spawn at the same time.
With the other options, the spawning of unison grain voices are offset into
the period before the next spawn, as follows:
X|Y Control
Y Axis Menu
Option Description
You can also adjust the pan and level of the signal.
Granular Parameters
Warp Mode
Setting the warp allows you to manipulate the
playback/sound of the sample.
You can also use the < > arrows to conveniently switch between different warp modes without having
to open the menu. After selecting a mode, you can use the knob to set the depth.
See “Exploring the Warp Modes” on page 49 for detailed information about the available warp
modes.
Scan
Use the SCAN knob to set the scan rate. The scan rate controls how quickly the Serum moves through
the audio sample to generate grains.
Granular synthesis involves breaking an audio sample into tiny pieces called grains (usually lasting a few
milliseconds) and then recombining them to create new sounds. Serum determines the playback start
position of a grain by the current playhead position.
The scan rate specifies how quickly the playhead moves through the sample. Setting the scan rate
higher spreads the start positions of grains out along the sample more, resulting in less overlap between
grains.
Setting the scan rate lower causes movement through the sample to slow down. This can produce a
more stretched, evolving, or drone-like sound.
In summary, a higher scan rate can be used to maintain rhythmic accuracy or to create fast-paced,
glitch-like effects. A slower scan rate allows for dramatic time-stretching effects, creating elongated and
ambient textures.
Setting the scan rate to a negative value reverses the direction of the playhead. Alternatively,
setting the scan rate to 0 stops the playhead from moving.
Density
Use the DENS knob to set the density of the grain cloud.
Density defines the rate at which grains are spawned, according to one of three options that you can
specify by right-clicking the knob and choosing from the context menu. The options are:
When selected, the context menu offers Triplet and Dotted as additional options, allowing you to
specify whether triplet or dotted divisions can be selected with the knob.
• Grains — The spawn rate is calculated as a function of grain length such that a consistent number
of grains, as set by the control, is playing at any given time
You can additionally specify two more options using the DENSITY knob context menu.
• Jump Start — Enable this option to have multiple grains spawn at note start so that the full density
is heard immediately.
When disabled, there is only a single spawning at note start and the sound builds to full density
over subsequent spawnings, giving the note a softer start.
• Max Grains — Set to place a limit on the maximum number of grains that can play at any one time,
including unison grain voices.
Many grains playing simultaneously can consume a lot of CPU. Use this option to help reduce this
consumption. If the oscillator tries to spawn a grain when the maximum number is playing, it will
skip and wait until one has stopped playing before it spawns again.
Length
Use the LENGTH knob to set the duration of each grain.
Setting the duration of each grain allows you to control how long each sound fragment lasts, influencing
the overall character of the resulting texture. Shorter grains can produce sharper, more rhythmic sounds,
while longer grains create smoother, more sustained tones.
You can specify how the length is determined by right-clicking the LENGTH knob and choosing one of
three options from the context menu:
• BPM Sync — Grain length can be set to a bar/beat division of host tempo
When selected, the context menu offers Triplet and Dotted as additional options, allowing you to
specify whether triplet or dotted divisions can be selected with the knob.
• Percent — Grain length is set to a percent of the density period. Note that this option is not
available if you select the Grains option with the DENSITY knob.
Pan
Use the PAN knob to control the placement of the sample in the stereo field (left to right).
Level
Use the LEVEL knob to control the output volume of the oscillator.
Grain Randomization
Knob Description
Grain Randomization
By altering specific frequency bands, adding or removing harmonics, or even shifting spectral content
over time, this approach can create complex, evolving textures that can range from natural acoustic-like
tones to entirely synthetic soundscapes. In practical use, spectral synthesis opens up unique possibilities
for morphing and transforming sound in ways that are not achievable with other synthesis techniques.
For instance, by isolating and processing specific frequencies, it is possible to create sounds that
gradually shift from one texture to another, or blend multiple sources into a single, coherent output.
Additionally, spectral synthesis allows for dynamic filtering and precise spectral editing.
Note: This method can be CPU intensive but provides unparalleled flexibility in shaping sound at a
fundamental, spectral level.
Spectral Synthesis
Spectral Menu
Spectral Waveform
Option Description
Smooth Apply a fourth-order Butterworth filter at the low and high frequency
boundaries, for smoother edges.
Post Warp Apply the low/high filtering after processing spectral warps.
Waveform Display
Loop Menu
Use the loop menu to specify the loop mode for
the sample.
It is not possible to drag the loop markers outside the playback start/end markers.
You can, however, drag, automate, or modulate the markers so that the loop end marker is
before the loop start marker. In this case, the loop direction is reversed.
Unison
Use the UNISON control to set the number of unison voices, effectively “stacking” oscillators in a way
that is similar to playing multiple notes of the same pitch, but slightly detuned.
Click the field and drag to set the appropriate value. You can also double-click the field and type a value.
Note: Unison causes Serum to generate multiple voices, raising CPU usage. The color of the UNISON
field changes as you increase the number of unison voices as a reminder of the CPU consumption.
Setting Description
This mode can sound very controlled and smooth, giving a thick and
coherent texture.
• Super — Multiple voices are slightly detuned from each other but with a
special emphasis on creating a dense and powerful sound, often with a
slight stereo spread.
Use this mode to create a lush, wide sound, especially with supersaw
sounds, where the detuned voices give a rich, full sonic character.
Setting Description
This mode can create unique, rich textures, especially for sounds that
need to be more aggressive or have a strong presence in a mix.
• Inv — The detuning behavior is inverted, with lower voices detuned
more sharply compared to higher ones.
This can create a more organic or chaotic sound, ideal for achieving
textures that are less polished and more natural or experimental.
STACK The unison stacking.
DETUNE The tuning offset for the additional voices.
BLEND The level offset of the unison voices versus the “central” unison voice or
voices (1 if an odd number, 2 if an even number of unison).
You can think of BLEND as a wet/dry mix between a unison (wet) and non-
unison (dry) sound. The default value of 75% is an even blend between all
the voices.
Note that this is only applicable when the number of unison voices is greater
than two.
WIDTH The extent to which the unison voices are spread out across the stereo field,
determining how wide or narrow the resulting sound feels in a stereo mix.
When you increase the width parameter, the unison voices are spread
farther apart in the stereo spectrum, placing some voices more to the left
and others more to the right. This creates a sense of space and a wider,
more immersive sound.
Conversely, decreasing the width narrows the stereo spread, making the
sound more centered and focused. This can be useful for sounds that need
to be more direct or fit better in a dense mix without dominating the stereo
field.
Setting Description
RANGE The extent or range of detuning applied to the unison voices, determining
how far apart in pitch the individual voices are spread around the central
frequency of the original sound.
When the range is set to a low value, the detuning between the unison
voices is minimal. The voices remain relatively close in pitch to the original
frequency, creating a subtle and smooth chorusing effect. This can add
warmth and a slight thickness to the sound without drastically changing its
character.
This creates subtle timing differences, which can make the sound more lively,
complex, and textured.
WARP 1 Spread out the warp amount applied to each voice around the current
WARP 1 knob position.
WARP 2 Spread out the warp amount applied to each voice around the current
WARP 2 knob position.
X|Y Control
Y Axis Menu
Option Description
You can also adjust the pan and level of the signal.
Spectral Parameters
Scan
Use the SCAN knob to set the speed and direction
of the sample playback.
Setting the sample length to BPM allows drum loops or samples to sync with the DAW tempo while still
letting you pitch the sample up and down using the keyboard while keeping the tempo consistent.
• Phase Lock — Adjust the FFT phases to minimize the audible phase change between FFT blocks
This can result in a less “smeared” sound, more faithful to the original sample. Consider using this
with tonal samples.
Cut
Use the CUT knob to set the cutoff of the spectral filter.
This sets the cutoff point of the spectral filter, determining which frequencies are allowed to pass
through or are filtered out.
By adjusting the CUT knob, you can control the range of frequencies that shape the sound, effectively
removing unwanted high or low spectral content to refine the tone and texture of the output.
Filter
You can create a custom filter curve, choose a filter preset, or choose a wavetable to act as the filter.
Filter Display
The following table describes operations you can perform when editing the filter:
Click and drag to select Select multiple points. This allows you to drag
multiple points simultaneously.
Click and drag in the GRID field to set the number of grid divisions. To set a specific value, double-click
the field and type the appropriate value.
Choosing a Wavetable
Filter Menu
Use the MIX knob to control the balance between the wet (processed) and dry (unprocessed) signal.
By default, warp is set to OFF (as displayed next to the corresponding knob). Clicking the current setting
displays a menu from which you can choose from among the available warp modes.
You can also use the < > arrows to conveniently switch between different warp modes without having
to open the menu. After selecting a mode, you can use the knob to set the depth.
See “Exploring the Warp Modes” on page 49 for detailed information about the available warp
modes.
Pan
Use the PAN knob to control the placement of the sample in the stereo field (left to right).
Level
Use the LEVEL knob to control the output volume of the oscillator.
Using the sub oscillator enhances bass presence by reinforcing low frequencies, giving your sound more
power and fullness. This is especially useful for creating bass-heavy sounds like sub-bass, deep bass for
electronic music, or warm, rich pads.
The sub oscillator also helps create a thicker, more robust tone, making it ideal for leads, plucks, and
pads, as it adds body without overwhelming the higher frequencies.
Sub Oscillator
Pitch
Use the OCT (octave) and CRS (coarse) controls to alter the pitch of the waveform.
The CRS setting controls the pitch transpose that tunes or detunes continuous (no snap) semitones.
CRS is most useful as a modulation destination or automation parameter.
Waveform
You can select the waveform to use as the basis of the sub oscillator. The following table describes the
waveform options:
Phase
Use the PHASE control to specify where the oscillator should begin playing back when a note is
triggered. This is similar to the phase control in other parts of Serum. It is also the same concept as
sample start on a sampler (except the sub oscillator “sample” is a very small waveform).
Right-click the PHASE control and choose Contiguous in the menu that appears to have new notes
continue with the phase of the previous note.
Pan
Use the PAN knob to control the placement of the waveform in the stereo field (left to right).
Level
Use the LEVEL knob to control the output volume of the sub oscillator.
Since the noise oscillator is actually a stereo sample player featuring high-quality playback, you can also
use it to load your own samples for even greater versatility.
For example, you could use the noise oscillator as a modulation source for many parameters (though the
results can, admittedly, be chaotic). Samples like drum loops, for example, can lead to very interesting
results.
The noise oscillator also appears in the WARP section of all three wavetable oscillators, allowing you to
apply frequency, phase, amplitude, and ring modulation (using the sampler as the modulator).
Noise Oscillator
When using the noise oscillator as a modulator, you should turn down the volume of the
noise oscillator (to just take advantage of the modulation effect).
Loading a Preset
To load a noise preset, click the current sample name and
choose an option in the menu that appears.
After choosing a preset, you can use the < > arrows to
conveniently switch between noise samples without having
to open the menu.
Noise Menu
The following table offers a brief description of the difference between these different noises:
White White noise has equal energy at all frequencies, meaning it covers the entire
audible spectrum (20 Hz to 20 kHz) with equal intensity.
It sounds like a constant hiss or the static between radio stations.
Because of its equal distribution across frequencies, white noise has a bright,
harsh sound.
Pink Pink noise has equal energy per octave, which means the energy decreases
as the frequency increases. Specifically, it has 3 dB less energy per octave as
the frequency doubles.
It has a warmer, less harsh sound compared to white noise, with more
emphasis on lower frequencies.
Brown Brown noise, also called Brownian noise or red noise, has energy that
decreases even more rapidly than pink noise, at 6 dB per octave.
It has a deep, rumbling sound, with much more emphasis on lower
frequencies and very little high-frequency content.
Geiger Geiger noise doesn’t have a uniform frequency spectrum like the others.
Instead, it is chaotic, consisting of random clicks or bursts that are not evenly
distributed across time or frequency.
It sounds like the clicks of a Geiger counter, hence the name. The noise is
made up of sporadic, random bursts with unpredictable intervals between
them.
Loading a Sample
Noise oscillator sounds are simply mono or stereo WAV files. This means that you can use your own
samples as noise sounds.
To load your own sample in the noise oscillator, click the current sample name to display the NOISE
menu and choose Load Sample. A dialog appears allowing you to locate the sample on your computer.
Alternatively, drag a sample from the Finder (macOS) or Explorer (Windows) to the oscillator waveform
pane.
You can embed a noise sample into your preset (when saving the preset) by displaying the NOISE menu
and choosing Embed in Preset.
This allows you to streamline your presets by packaging your noise samples together with the Serum
settings.
Noise oscillator playback uses high-quality real time interpolation. This is because with
noise sounds, the high-frequencies are important. Serum strives to offer the best quality
whenever possible, even at the expense of additional CPU usage (although this feature is
heavily optimized with both SSE2 and pre-calculations when a file loads).
Since high-quality playback does require more CPU processing, it is worth noting that
mono sounds do use slightly less CPU resources (one channel instead of two can add up
with chords). The main consideration here is that if you have a mono noise source, don’t
export it as a stereo file for Serum since you will be wasting both disk space and CPU
resources (on playback).
One Shot/Looping
By default, the noise oscillator is set to loop samples.
Select the (one shot) switch to have the sample (noise) stop when playback reaches the end
of the sound file.
Looping is typically good with noise sounds, but one-shot mode is useful for attack sounds (to add a
percussive punch/attack transient to a sound, for instance).
Start
Use the START control to set the phase start (as a percentage). Click and drag in the field to set the
value. Alternatively, double-click the field and type the appropriate value.
This setting for the noise oscillator is similar to the phase controls in other parts of Serum. However,
since a noise sound is a lot longer than an oscillator waveform, it might be easier to think of this as
“sample start.”
You can also automate the control, which results in a sort of lo-fi “scratching” effect.
Random
Use the RAND control to randomize the start phase. This prevents the noise from being identical each
time you press a note.
Holding a chord is a good example of when you would likely want to add randomness, otherwise the
same noise sounds for all the notes.
Pitch
Use the PITCH knob to specify the base pitch/frequency for the noise oscillator. The default (50%) is
nominal pitch, that is, the original pitch of the sound file (assuming that the keytrack switch is off).
Fine
Use the FINE knob to fine tune the pitch setting.
Pan
Use the PAN knob to control the placement of the noise in the stereo field (left to right).
Level
Use the LEVEL knob to control the output volume of the noise oscillator.
By controlling which parts of the sound’s spectrum are allowed to pass through and which are cut off,
you can use Serum’s filters to drastically alter a sound’s character, making it warmer, brighter, darker, or
even more aggressive.
Serum includes an extensive collection of filters, featuring numerous variations on low-pass, high-pass,
band-pass, notch, comb, and more. Combined with Serum’s extensive modulation capabilities, you can
use the filters to create a range of sounds including rhythmic sweeps, evolving pads, and even vocal-like
formant sounds.
With playful tweaking and clever modulation, you can use the filter module as a creative playground,
exploring endless possibilities for sound manipulation and transformation.
Routing an Oscillator
To route an oscillator to a specific filter module, select the
corresponding switch:
• S - SUB oscillator
• A - OSC A
• B - OSC B
• C - OSC C
• N - NOISE oscillator
Filter Routing
Filter Type
To choose a filter type, click the current filter setting. A
pop-up menu appears showing a hierarchical collection of
available filter types.
Alternatively, hover over the menu and use the mouse wheel
to quickly rotate through menu options.
Filter Menu
Normal (cont.) High 6/12/18/24 State-Variable High-Pass FAT: Add saturation to the
Filter. filter resonance signal path.
The number represents This has a taming effect on
the db per octave slope of filter resonance and enriches
the filter. the resulting harmonic
content.
Band/Peak/Notch State-Variable Band/Peak/ FAT: Add saturation to the
12/24 Notch Filter. filter resonance signal path.
This has a taming effect on
filter resonance and enriches
the resulting harmonic
content.
Multi LH/LB/LP/LN/ Dual SVF Filters. FREQ: Set the cutoff
HB/HP/HN/BP/ frequency of the second SVF
BN/PP/PN/NN The first letter is primary,
filter.
the second letter is
secondary (for example,
BP is Band+Peak).
The resonance control
applies equally to both
filters.
LBH/LPH/LNH/ Morphing SVF filters (for MORPH: Smoothly transition
BPN example, Lowpass <-> between the filter states.
Bandpass <-> Highpass)
Flanges Cmb L/Flg L/Phs L Comb/Flanger/Phaser LP FREQ: Set the cutoff
with a low-pass filter in frequency of the low pass
the internal feedback filter affecting the internal
circuit. feedback circuit.
Set the MIX knob to 50%
for best results.
Cmb H/Flg H/Phs Comb/Flanger/Phaser HP FREQ: Set the cutoff
H with a high-pass filter frequency of the high-pass
in the internal feedback filter affecting the internal
circuit. feedback circuit.
Set the MIX knob to 50%
for best results.
Cmb HL/Flg HL/ Comb/Flanger/Phaser HL WID: Expand the
Phs HL with a high pass + low bandwidth allowed through
pass filter in the internal the internal feedback circuit
feedback circuit. around the filter cutoff
frequency.
Set the MIX knob to 50%
for best results.
Misc (cont.) Formant-I/II/III Formant ‘vowel’ filters. FORMNT: Shift the formants
to generate a broader
The CUTOFF knob
range of possible filter
morphs between various
permutations.
formants. This is great
for adding vocal-like
characteristics to patches.
Bandreject Attenuate a specific range WIDTH: Adjust the
to very low levels. It has overall width of the filter’s
the opposite effect of a attenuated “notch”.
band-pass filter.
Dist.Comb 1/2 LP/ Combination comb filter COMBFRQ: Set the
BP and pass filter. frequency of the comb filter.
New (cont.) Diffusor All-pass diffusor stage STAGES: Specify how many
for making things sound APF stages are used.
phasey and blurry in a
cool way.
MG Ladder Clean circuit model of SMOOTH: Slow the rate
classic transistor ladder at which the cutoff is
VCF. modulated.
Acid Ladder Circuit model of a diode SMOOTH: Slow the rate
ladder VCF ubiquitous in at which the cutoff is
acid music. modulated.
EMS Ladder Clean circuit model of the SMOOTH: Slow the rate
VCF that makes Dr. Who at which the cutoff is
sounds. modulated.
MG Dirty MG Ladder, but all the PAIN: How far you’re holding
distortion is there and you a lighter from the circuit
are overdriving the circuit. board (this is physically
correct; not a joke).
PZ SVF Drawable filters. SMOOTH: Slow the rate
at which the cutoff is
modulated.
Comb 2 Comb filter with crazy FRQ2
resonance and stuff.
Exp MM Multimode (LPF/Notch/ MIX: blends between LPF/
HPF) output from the Notch/HPF.
classic synthesizer
expander module.
Exp BPF BPF output from the N/A
classic synthesizer
expander module.
You can quickly cycle through the different display modes by Option-clicking (macOS) or Alt-
clicking (Windows) in the filter display.
Cutoff
Use the CUTOFF knob to set the primary cutoff frequency for the filter (with just a couple exceptions,
such as vowels for formant filters).
Use the (keytrack) switch to offset the cutoff using MIDI notes. With most filter types, one
octave of MIDI corresponds to precisely one octave of filter frequency control.
If the keytrack switch is enabled, this tracks the pitch of the first oscillator (OSC A, OSC B,
or OSC C) that has pitch tracking enabled (including portamento).
If none of the oscillators have pitch tracking enabled, the filter tracks the input MIDI note
number.
Resonance
Use the RES knob to set the resonance (feedback) of
the filter circuit.
Graphical Adjustments
Drive
Use the DRIVE knob to increase the gain into the filter
circuit and can impart some coloration (mild distortion)
to the sound.
Refer to the table above for details about the knob that appears here.
Pan
Use the PAN knob to create a cutoff offset for the left and right signals. At the default setting of 50%
(12 o’clock) this knob has no effect. When turned to the left (counter-clockwise) the left channel cutoff
increases, and the right channel cutoff decreases.
When turned to the right (past 12 o’clock clockwise), the opposite happens; the left cutoff decreases
and the right channel cutoff increases.
Mix
Use the MIX knob to control the wet/dry amount for the filter. The default (100%) means 100% wet.
Level
Use the LEVEL slider to adjust the filter output level (in decibels).
In short, the mixer provides you with precise control over the individual elements of your sound. Using
the Serum mixer, you can create complex, layered sounds with ease, adjusting the volume, pan, and
routing of each component to achieve the perfect sound.
The mixer shows the elements of Serum that you have enabled.
For the Init patch, for instance, only OSC A is enabled and, therefore, this is the only channel enabled in
the mixer. Click the header to enable a channel.
Hold the Option/Alt key and drag the oscillator channel label to another oscillator label to
copy the oscillator without modulations.
Hold the Shift-Option or Shift-Alt keys and drag the oscillator channel label to another
oscillator label to copy with modulations.
Dragging from one oscillator channel label to another without any keyboard modifiers swaps
the two oscillators (including modulation assignments).
Option Description
Filter Route the signal to the filter module. Use the top knob to set how much of
the signal is sent to FILTER 1 and FILTER 2.
Main Route the signal to the main output. When you choose this option, the
envelope button appears near the bottom left of the channel.
When enabled, the oscillator output level is affected by ENV 1. When
disabled, ENV 1 has no effect on the oscillator output level and it terminates
after the longest release of any of the envelopes. You can use this to make
oscillators appear closer to being free running.
Direct Route the signal to direct output, bypassing the filter and effects sections,
outputting “clean” along with the main output.
Similar to MAIN, when you choose this option, the envelope button
appears near the bottom left of the channel.
None Disable the output of the source (the oscillator, for example).
Sending a signal through auxiliary effects busses offers multiple advantages, including the following:
Using a shared bus for effects such as reverb or delay can create a more cohesive sound across the
mix. For example, sending multiple tracks to a single reverb bus can make them sound as if they’re
in the same space, creating a sense of depth and unity.
Auxiliary busses allow you to control the wet/dry balance more precisely, as you can adjust the
send levels from each track to the effect bus. This flexibility makes it easier to blend the effect
subtly or aggressively, depending on the needs of the mix.
• Parallel processing
Aux busses enable parallel processing, where the dry signal and the processed signal run
simultaneously. This is useful for effects like parallel compression, where you maintain the original
dynamics while adding enhanced processing to the signal.
• CPU efficiency
Instead of applying the same effect to multiple oscillators individually, you can send them to a
single auxiliary bus. This reduces the CPU load since only one instance of the effect is running.
Click and drag the PAN knob left or right, as appropriate. To set
a specific value, double-click the knob and type the appropriate
value in the field that appears.
Use negative values for left and positive values for right.
You can use panning to effect the following: Setting Pan and Levels
• Create a stereo image — Panning places sound elements across the stereo field (left to right), giving
the mix width and spatial dimension.
By positioning elements in different parts of the stereo field, you create a more realistic and
immersive sound, emulating how we naturally hear sounds around us.
• Prevent clashing — Panning helps separate elements that occupy similar frequency ranges.
For example, if you have two midrange-heavy sound elements, panning them apart reduces the
likelihood of them competing for space and allows each to be heard more clearly.
• Add depth and realism — Proper use of panning mimics the natural placement of sound elements
in a room or on stage. It helps give the listener a sense of depth and positioning, making the mix
feel more dynamic and engaging.
• Ensure balance — Adjusting levels is essential for achieving a clear and balanced mix.
Setting the volume of each element ensures that the most important parts are prominent while
supporting elements are appropriately audible without overpowering the mix.
• Maintain dynamics — Setting levels effectively preserves the dynamic relationships between sound
elements.
By raising or lowering the volume of certain elements, you can guide the listener’s attention and
shape the overall feel of the sound, making certain parts stand out more prominently while others
remain in the background.
Mixing Filters
You can control how the filter output
integrates into the overall mix.
Similar to the filter module, you can graphically adjust the filter cutoff and
resonance (in combination) by clicking and dragging in the filter display.
Filter Adjustments
The following table describes the signal routing options from the filters:
Option Description
Filter 1 or Filter 2 Route the signal to the other filter, either FILTER 1 or FILTER 2.
Main Route the signal to the main output. When you choose this option, the
envelope button appears near the bottom left of the channel.
When enabled, the oscillator output level is affected by ENV 1. When
disabled, ENV 1 has no effect on the oscillator output level and it terminates
after the longest release of any of the envelopes. You can use this to make
filters appear closer to being free running.
Direct Route the signal to direct output. Similar to MAIN, when you choose this
option, the envelope button appears near the bottom left of the
channel.
None Disable the output of the source (the filter, for example).
Similarly, click and drag the MIX knob to set the dry/wet
blend of the filter mix.
Note: The MIX knob has no effect for Combs-type filters. Setting Pan, MIX and Levels
Finally, click and drag the level up or down, as needed. To set a specific value, double-click the control
and type the appropriate value in the field that appears.
The following table describes the signal routing options from the busses:
Option Description
Setting Levels
You can set the level for each bus.
Setting Levels
An empty rack appears that you can populate with any of the 13 effects modules, in any order.
Click the button (near the top left) to expand the FX rack and list view. Alternatively, press
Option-F (macOS)/Alt-F (Windows) to expand the view.
Click the button to revert the FX rack (and list view) back to its original size. Similarly, press
Option-F (macOS)/Alt-F (Windows) to revert the view to the original size.
Selecting a Rack
Serum offers three FX racks: MAIN, BUS 1,
and BUS 2. Each FX rack processes the audio
signal on the corresponding channel.
Serum FX Racks
FX Presets Menu
Adding Modules
Click the button and choose a module in the list
that appears.
Adding an FX Module
Reordering Modules
To reorder effects, click and drag an effect to the new location in the list view (on the left) or in the rack
view (on the right). A yellow line indicates where the module will land.
Reordering Modules
Copying a Module
You can copy an FX module to create a duplicate of the module on a rack, with or without assigned
modulations.
To copy a module without modulations, Option-drag (macOS) or Alt-drag (Windows) an existing module
to the appropriate location on the rack. A yellow line indicates where the module will land, which is
helpful when placing a module between two existing modules.
Copying a Module
Bypassing a Module
Every FX module features a bypass button that allows you to easily bypass the module. This
button appears to the right of the module both in the list and rack view.
Clicking the button bypasses the module in the signal routing. When enabled, the button highlights in
red to show that the module is being bypassed.
Bypassing a module is really intended for temporary use. For example, you can use bypass to hear your
sound with and without a given FX (without having to set the MIX knob to 100% dry).
Option-clicking (macOS) or Alt-clicking (Windows) on a bypass button toggles bypass for all FX
on the bus. This also works with the FX bypass buttons on the MIXER page.
Removing a Module
To deactivate an effect, you should generally
remove the module from the rack (you can always
add it back later).
Remove an FX Module
Click the button. A dialog appears allowing you to type the rack preset name. By default, the preset
is saved in a standard user location so that Serum can easily find it later.
Operation Description
Modulating FX Parameters
You can modulate most FX parameters, similar to the way you can modulate standard Serum synth
controls.
The FX rack is a DSP process that operates on the sum output of the synth engine
(rather than per voice). You can think of this as “the effects are monophonic” or “the
effects are like plug-in inserts after Serum.” This is sometimes referred to as paraphonic
behavior, especially in conjunction with the filter effect.
Therefore when playing polyphonic synth parts (strummed chords, for example) keep
in mind that automating FX controls with per-voice mod sources (such as an envelope)
results in the effect parameter modulation being modulated/retriggered by each new
note.
Click the button to display the module menu. Note that the color of the button varies depending on
the color of the module. A drop-down menu appears.
Module Menu
Operation Description
Bode
The BODE module offers an implementation of the Bode frequency shifter, a device that shifts the
frequency of an audio signal by a fixed amount, resulting in a unique sound effect.
Named after electronic music and audio signal processing pioneer Harald Bode, this frequency shifting
can create dissonance, phasing effects, or a sense of movement in sound, which you can creatively
apply to your sound design.
Bode Module
Control Description
Chorus
The CHORUS module offers a four-voice chorus effect, with two left and two right chorus taps.
Chorus Module
Control Description
RATE The rate of the chorus. The units depend on the BPM setting. When BPM is
on, the RATE knob snaps to musical time (from 8 bars to 1/32nd note).
When BPM is off, the RATE knob is in Hertz (Hz), between 0 Hz and 20 Hz.
BPM Specify whether the chorus is synced to the host BPM. See the RATE knob
(above) for more information.
DELAY 1 The amount of delay (in milliseconds) between the dry signal and the first
stereo pair of chorus voices.
DELAY 2 The amount of delay (in milliseconds) between the dry signal and the second
pair of chorus voices.
DEPTH Specifies how much the chorus LFO modulates the delay times described
above (how much pitch warble occurs).
FEEDBACK The feedback amount of the chorus voices (how much of the chorus voice
output appears back at the input of the chorus module). This creates a more
pronounced “ringing” to the chorus.
LPF/HPF The cutoff frequency in Hertz (Hz) of the low pass/high pass filter after the
chorus wet effect. This is useful for a more (or less) “warm doubling” of the
signal. Click the label to toggle between LPF and HPF.
MIX The wet/dry amount for the effect, from 0 (100% dry) to 100 (100% wet).
LEVEL The output level of the module (in decibels).
Compressor
The COMPRESSOR module reduces the volume of loud sounds or amplifies quiet sounds, thereby
reducing or compressing the dynamic range of an audio signal.
Compressor Module
Control Description
Control Description
ATTACK The amount of time in milliseconds for the gain reduction to engage.
A longer (slow) attack is useful for letting some signal through before the
gain reduction takes place, resulting in a “punch,” “snap,” or “bite” (sounds are
hard to describe using language!).
A shorter/faster attack tames peaks more completely.
RELEASE The amount of time for the gain reduction to be removed.
GAIN The amount of makeup gain. This is a good way to boost quiet signals. The
control allows for approximately 30dB of boost for the compressor and
36dB for the limiter so be careful (that’s a large amount of gain). Basically, a
little can go a long way.
X-LOW (Enabled when MULTIBAND is selected) Sets the low crossover for the
multiband split.
BELOW (Enabled when MULTIBAND is selected) Sets the compression ratio below
the threshold.
X-HIGH (Enabled when MULTIBAND is selected) Sets the high crossover for the
multiband split.
H (Enabled when MULTIBAND is selected) Sets the high band gain.
M (Enabled when MULTIBAND is selected) Sets the mid band gain.
L (Enabled when MULTIBAND is selected) Sets the low band gain.
MIX The wet/dry amount for the effect, from 0 (100% dry) to 100 (100% wet).
LEVEL The output level of the module (in decibels).
Convolve
The CONVOLVE module allows you to apply the impulse response of a signal you select (representing
the characteristics of a room’s reverb or a specific filter) to your sound design.
The most common use is to create a convolution reverb, where the impulse response of a real acoustic
space (such as a concert hall) is convoluted with your audio signal to simulate how that signal would
sound if played in that space. This allows for highly realistic and complex reverb effects.
You can also use the CONVOLVE module to blend sounds in unique ways, applying an effect that alters
your sound in a manner different from simple filtering or other modulation techniques.
Convolve Module
Control Description
IMPULSE Click to choose from a menu of impulse responses. Use the < > arrows to
advance through the modes without having to open the menu.
You can load impulse responses from outside the Serum 2 Presets
folder by dragging and dropping the files on the IR display or selecting Load
IR from the context menu.
If you do this, the option to Embed in Preset appears in the menu, and an
embed icon appears at the top right of the IR display. This allows you to
embed the impulse response into the preset, similar to what you can do with
oscillators.
SIZE The size with which to stretch or contract the impulse.
TONE Use to filter the impulse.
ϕ MIN The convolution minimum phase. This converts the IR to a minimum-phase
representation, keeping the frequency response unchanged and eliminating
echoing.
PRE-DLY The convolution pre-delay. This offsets the impulse in time.
BPM Specifies whether the pre-delay is synced (BPM or millisecond based.
ATTACK The convolution attack. Use this to fade in the impulse.
DECAY The convolution decay. Use to shorten the impulse.
DAMP Use to shorten the high frequency of the impulse.
IR GAIN The impulse volume.
MIX The wet/dry amount for the effect, from 0 (100% dry) to 100 (100% wet).
LEVEL The output level of the module (in decibels).
Delay
The DELAY module records an input signal and then plays it back after a period of time. The delayed
signal may be played back multiple times, or fed back into the recording, to create the sound of a
repeating, decaying echo.
Delay Module
Control Description
In addition to using the FREQ and Q knobs, you can manipulate the delay filter using your mouse.
Distortion
The DISTORTION module offers 13 types of distortion, including two dual-waveshaper modes that
allow you to create your own custom distortion.
Distortion Module
Control Description
MODE Click to choose from a menu of distortion types (Tube, by default). You can
use the < > arrows to advance through the modes without having to open
the menu.
OFF/PRE/POST A switch to enable filtering, which you can set to either pre-distortion (PRE)
or post-distortion (POST).
TYPE The filter type for the distortion module. Drag the red control between left
and right to morph from low pass to bandpass to high pass.
FREQ The cutoff frequency for the filter (when the filter is enabled). Double-click
the field to display a text box, allowing you to enter a frequency value.
Right-click the knob and choose Key Track in the context menu to have
frequency respond to the pitch of the note played. With Key Track disabled,
the frequency is fixed regardless of the key played.
With Key Track enabled, the frequency changes in proportion to the pitch
of the note played (higher-pitched notes increase the frequency).
Q The resonance for the filter (when the filter is enabled). Double-click the
field to display a text box, allowing you to enter a resonance value.
You can set a high value to create squelchy feedback. Typically, lower values
are more common.
DRIVE Generally, the gain boost for the distortion, with the following exceptions:
• With Downsample filter type, the DRIVE knob controls the sample rate
reduction amount.
• With X-Shaper and X-Shaper (Asym) modes, the DRIVE knob affects a
morph between the two waveshapes.
MIX The wet/dry amount for the effect, from 0 (100% dry) to 100 (100% wet).
LEVEL The output level of the module (in decibels).
The X-Shaper is a dual crossfading waveshaper. Selecting X-Shaper in the distortion menu causes Edit
A and Edit B buttons to appear directly below the menu.
Clicking either button displays a pop-up “X-Y” graph editor for the respective waveshaper. In both cases,
the X (horizontal) axis represents the input level, and the Y (vertical) axis represents the corresponding
remapped output level for an input level.
The DRIVE knob, described above, controls the blend between the two waveshaping graphs (the
DRIVE knob at 0% presents waveshaper A while 100% presents waveshaper B).
Note that X-Shaper is a symmetric waveshaper, with the lower-left point on the graph representing
silence (-INF dB for input and output). Similarly, the top-right point represents the highest level (0 dB for
input and output).
In contrast, X-shaper (Asym) is an asymmetric waveshaper. In this case, the middle of the graph
represents silence (-INF dB input and output), the top-right represents the highest positive value to the
signal, and the lower-left represents the highest possible negative value.
Asymmetric distortion allows you to bring out even-order harmonics that are not typically found in a
standard symmetric distortion (such as clipping). This is often the case in guitar amps; one pole distorts
(for example, fatline) while the other pole remains relatively undistorted.
Equalizer
The EQUALIZER module offers two-band parametric control.
Equalizer Module
You can set the type of each of the two bands using the corresponding three-state switches. The left
band offers low-frequency (LF) adjustment and enables low shelf, peaking, or high pass filtering. The
right band offers high-frequency (HF) adjustment and enables high shelf, peaking, or low pass filtering.
Control Description
FREQ (L) The frequency (in Hz) for the low EQ band.
Q (L) The Q (resonance) for the low EQ band.
GAIN (L) The gain boost/cut (in dB) for the low EQ band. This knob has no effect if
you selected High Pass as the low EQ band type.
Control Description
FILTER TYPE Click the icons to select a Shelf, Peak, or High Pass filter type (on the left)
and Shelf, Peak, or Low Pass filter type (on the right).
FREQ (R) The frequency (in Hz) for the high EQ band.
Q (R) The Q (resonance) for the high EQ band.
GAIN (R) The gain boost/cut (in dB) for the high EQ band. This knob has no effect if
you selected Lowpass as the high EQ band type.
LEVEL The output level of the module (in decibels).
Filter
The FILTER module operates identically to the per-voice synth filter found on the main OSC tab, except
that in this case, it runs as a master effect.
Filter Module
Control Description
TYPE Click to choose from a menu of filter types (MG Low 6, by default). You can
use the < > arrows to advance through the filter types without having to
open the menu.
See “Filter Types and Var Parameter Functions” for more information about
the available filter types.
CUTOFF The primary cutoff frequency for the filter (with just a couple exceptions,
such as vowels for formant filters).
TIP: You can recreate certain paraphonic vintage synth behaviors by
applying an envelope to this knob.
Right-click the knob and choose Key Track in the context menu to have
cutoff frequency respond to the pitch of the note played. With Key Track
disabled, the cutoff frequency is fixed regardless of the key played.
With Key Track enabled, the cutoff frequency changes in proportion to the
pitch of the note played (higher-pitched notes increase the cutoff frequency.
RES The resonance (feedback) of the filter circuit.
Control Description
DRIVE The gain into the filter circuit. The setting can impart some coloration (mild
distortion) to the sound.
Right-click the knob and choose Clean Mode in the context menu to have
the filter pre-gain stage the filter input -24 dB (with a +24 dB boost post-
filter).
FAT/FREQ/MORPH/ This is a variable knob with different functions depending on the selected
LP FRO/HP FRO/ filter type.
HL WID/LP FRO/
DB +/-/SPREAD/ For instance, with “dual” filters, the VAR knob controls the second filter
DAMP/BOEUF/ cutoff frequency.
THRU/FORMNT/
WIDTH/COMBFRO/
SCREAM/STAGES/
SMOOTH/PAIN/
FRO2
PAN A cutoff offset for the left and right signals. At the default setting of 50% (12
o’clock) this knob has no effect. When turned to the left (counter-clockwise)
the left channel cutoff increases, and the right channel cutoff decreases.
When turned to the right (past 12 o’clock clockwise), the opposite happens;
the left cutoff decreases and the right channel cutoff increases.
MIX The wet/dry amount for the effect, from 0 (100% dry) to 100 (100% wet).
LEVEL The output level of the module (in decibels).
You can graphically adjust the filter cutoff and resonance (in
combination) by clicking and dragging in the filter display.
Right-click in the display to access a menu allowing you to choose Graphical Adjustments
display options for the filter.
You can quickly cycle through the different display modes by Option-clicking (macOS) or Alt-
clicking (Windows) in the filter display.
Flanger
The FLANGER module works by cyclically varying phase shift into one of two identical copies of a signal
and then recombining them.
Flanger Module
Control Description
RATE The rate of the flanger. The units depend on the BPM setting. When BPM
is on, the RATE knob snaps to musical time (from 8 bars to 1/32nd note).
When BPM is off, the RATE knob is in Hertz (Hz), between 0 Hz and 20 Hz.
BPM Specifies whether the flanger sweep is synced to the host BPM. See the
RATE knob (above) for more information.
DEPTH Specifies how much the flanger LFO influences the sound, in other words,
how much (or deep) the flanger operates.
FEEDBACK The feedback amount of the flanger circuit, which makes the effect more
pronounced (“ringing”).
Control Description
PHASE The stereo phase offset for the LFO influence over the flanger (the left
flange and right flange offset). A setting of 0% sets both left and right to the
same frequency.
A setting of 50% represents 180 degrees, meaning that the left and right
have opposite frequencies. In this case, the flanger sweep “rises” on the left
while “falling” on the right, or vice versa.
MIX The wet/dry amount for the effect, from 0 (100% dry) to 100 (100% wet).
LEVEL The output level of the module (in decibels).
Hyper/Dimension
The HYPER/DIMENSION module is a micro-delay chorus with a variable number of voices (1-7). In
addition you can configure the Hyper/Dimension effect to retrigger on every MIDI note, which adds to
the potential simulation of a unison.
To conserve CPU, consider using the HYPER effect as an alternative to high unison settings.
Hyper/Dimension Module
Control Description
RATE The speed at which the various hyper voices oscillate sharp/fat in pitch.
UNISON The number of chorus voices. If you only want to use the DIMENSION
effect and not the HYPER effect, set the UNISON to 0.
DETUNE The amount/depth of the hyper voice oscillations (sharp/fat in pitch).
RETRIG When turned on, resets all hyper voices to start over from a zeroed-pitch
offset.
This provides a laser-like zap effect for each note-on event. You might use
this on certain monophonic patches, for example.
MIX The wet/dry amount for the effect, from 0 (100% dry) to 100 (100% wet).
LEVEL The output level of the module (in decibels).
The DIMENSION effect is a pseudo-stereo effect consisting of four delay lines summed out-of-phase
and slowly amplitude-modulated to provide a subtle amount of motion to the effect. This is useful for
adding a perceived width to an otherwise mono signal.
Control Description
Phaser
The PHASER module filters a signal by creating a series of peaks and troughs in the frequency
spectrum.
Phaser Module
Control Description
RATE The rate of the phaser. The units depend on the BPM setting. When BPM
is on, the RATE knob snaps to musical time (from 8 bars to 1/32nd note).
When BPM is off, the RATE knob is in Hertz (Hz), between 0 Hz and 20 Hz.
BPM Specifies whether the phaser sweep is synced to the host BPM. See the
RATE knob (above) for more information.
POLES The number of stacked phaser poles.
DEPTH Specifies how much the phaser LFO influences the sound.
DEPTH 2 The offset between phaser stages.
FREQ The base frequency for the phaser effect.
FEEDBACK The feedback amount of the phaser circuit. A higher setting makes the effect
more pronounced (“ringing”).
PHASE The stereo phase offset for the LFO influence over the phaser (left flange
and right flange offset). A setting of 0% sets both left and right to the same
frequency.
A setting of 50% represents 180 degrees, with left and right at opposite
frequencies (in other words, the phaser sweep “rises” on the left while
“falling” on the right, or vice versa.
Control Description
MIX The wet/dry amount for the effect, from 0 (100% dry) to 100 (100% wet).
LEVEL The output level of the module (in decibels).
Reverb
The REVERB module offers a plate and hall reverb, using a modified version of the Tal Reverb algorithm
(courtesy of Togu Audio Line).
Reverb Module
• Space
• Marble
• Rectangle
• Hexagon
• Box
NITROUS CHORUS The top value sets the speed that the reverb is modulated.
(cont.)
The bottom value sets the pitch depth that the reverb is
modulated.
BASIN LO CUT Suppress low frequencies from the reverb. 0% means no effect
on lows, 100% means no lows at all.
HI CUT Suppress high frequencies from the reverb. 0% means no effect
on highs, 100% means no high frequencies at all.
SIZE The length of the reverb.
PRE-DLY The offset time for the reverb.
FEEDBACK The amount of reverb fed back into the input signal.
CHORUS The top value sets the speed that the reverb is modulated.
The bottom value sets the pitch depth that the reverb is
modulated.
MIX The wet/dry amount for the effect, from 0 (100% dry) to 100
(100% wet).
LEVEL The output level of the module (in decibels).
Splitter L/H
The SPLITTER L/H module divides the audio signal into distinct low and high-frequency bands,
enabling you to design and apply dedicated FX racks for each band independently.
This setup provides precise control over your sound, allowing for tailored processing that enhances both
the low-end punch and high-end clarity.
When using this module, you can build two separate racks, one to handle the LOWS and another to
process the HIGHS. The list view (on the left) shows both racks. The rack view, however, only shows the
currently-selected rack. Click the LOWS or HIGHS panel to display the corresponding rack.
Control Description
LOWS Right-click in the panel and choose an FX module to add using the menu
that appears. Continue doing this until you finish building the low-end rack.
Each time you add a module, it appears below the SPLITTER L/H main
module.
Note the bypass button in the LOWS panel. Enabling this bypasses the low-
end rack.
SPLIT FREQ The crossover frequency for the low and high bands.
HIGHS Right-click in the panel and choose an FX module to add using the menu
that appears. Continue doing this until you finish building the high-end rack.
Each time you add a module, it appears below the SPLITTER L/H main
module.
Note the bypass button in the HIGHS panel. Enabling this bypasses the
high-end rack.
LEVEL The output level of the module (in decibels).
Splitter L/M/H
The SPLITTER L/M/H module divides the audio signal into low, mid, and high-frequency bands,
offering even greater flexibility than the SPLITTER L/H module. This configuration enables you to
design and apply dedicated FX racks for each band independently, allowing for fine-tuned control over
your sound.
With the addition of the MIDS panel, you gain the ability to shape and process the critical midrange
frequencies separately, ensuring that elements like vocals, guitars, and synths stand out or blend
seamlessly. This setup provides precise control, enhancing the low-end punch, midrange presence, and
high-end clarity, for a more refined and impactful mix.
When using this module, you can build three distinct racks, one to handle the LOWS, one for the MIDS,
and another to process the HIGHS. The list view (on the left) shows all three racks. The rack view,
however, only shows the currently-selected rack. Click the LOWS, MIDS, or HIGHS panel to display the
corresponding rack.
Control Description
LOWS Right-click in the panel and choose an FX module to add using the menu
that appears. Continue doing this until you finish building the low-end rack.
Each time you add a module, it appears below the SPLITTER L/M/H main
module.
Note the bypass button in the LOWS panel. Enabling this bypasses the low-
end rack.
Control Description
SPLIT FREQ The crossover frequency for the low and mid bands.
MIDS Right-click in the panel and choose an FX module to add using the menu
that appears. Continue doing this until you finish building the low-end rack.
Each time you add a module, it appears below the SPLITTER L/M/H main
module.
Note the bypass button in the MIDS panel. Enabling this bypasses the mid-
end rack.
SPLIT FREQ The crossover frequency for the mid and high bands.
HIGHS Right-click in the panel and choose an FX module to add using the menu
that appears. Continue doing this until you finish building the high-end rack.
Each time you add a module, it appears below the SPLITTER L/M/H main
module.
Note the bypass button in the HIGHS panel. Enabling this bypasses the
high-end rack.
LEVEL The output level of the module (in decibels).
Splitter MS
The SPLITTER M/S module divides the audio signal into mid and side bands. This configuration enables
you to design and apply dedicated FX racks for each band independently, allowing for precise control
over the central and spatial elements of your sound.
Use it to enhance focus and presence by processing the mid channel, or to control depth and width by
processing the side channel, giving you a powerful tool for shaping your mix.
When using this module, you can build two separate racks, one to handle the MID band and another to
process the SIDE band. The list view (on the left) shows both racks. The rack view, however, only shows
the currently-selected rack. Click the MID or SIDE panel to display the corresponding rack.
Control Description
MID Right-click in the panel and choose an FX module to add using the menu
that appears. Continue doing this until you finish building the mid-band rack.
Each time you add a module, it appears below the SPLITTER M/S main
module.
Note the bypass button in the MID panel. Enabling this bypasses the mid-
band rack.
SIDE Right-click in the panel and choose an FX module to add using the menu
that appears. Continue doing this until you finish building the side-band
rack.
Each time you add a module, it appears below the SPLITTER M/S main
module.
Note the bypass button in the SIDE panel. Enabling this bypasses the side-
band rack.
LEVEL The output level of the module (in decibels).
Utility
The UTILITY module offers a series of “utility” functions including polarity inversion, basic low and high
pass filters, stereo width and balance, and more.
Utility Module
Control Description
POLARITY INV Invert the polarity of the audio signal on the left and right channel of the
signal respectively.
LPF The low pass filter.
HPF The high pass filter.
MONO BASS/FREQ Enable for mono bass, which forces frequencies below the threshold (set
using the FREQ control) to be monophonic.
WIDTH The stereo width.
PAN The stereo balance.
MIX The wet/dry amount for the effect, from 0 (100% dry) to 100 (100% wet).
LEVEL The output level of the module (in decibels).
1. Envelopes
2. LFOs
3. Modulation Matrix
Using Envelopes
The Envelopes module offers four modulation sources, labeled ENV 1, ENV 2, ENV 3, and ENV 4. The
following shows the Envelopes area of Serum.
Serum Envelopes
Configuring Envelopes
Select an envelope by clicking the
corresponding envelope tab.
Locked The envelope always zooms to perfectly fill the display area. This means that
adjusting envelope times simply changes how much time is represented on
the ruler scaling below the envelope.
Unlocked Allows you to zoom in and out of the envelope by dragging up and down in
the zoom area directly below the Lock button.
See “Modifying Envelopes” later in this chapter for details about changing envelope shapes.
You can use all four envelopes, though ENV 1 is considered special within Serum because it is
used with the amplifier, controlling the output volume of each voice. You can still use ENV 1,
however, assigning it to any parameter as you would the other envelopes.
When you select ENV 2, ENV 3, or ENV 4, the envelope of ENV 1 is faintly visible (in gray) in
the background.
When the main LEGATO switch is enabled, envelopes do not retrigger if a second note is played while
a first note is still held. You can invert this option to force an envelope to always trigger at note on, even
when legato is enabled.
To do this, right-click the envelope graph and choose Legato Inverted in the context menu.
You can choose to alter values in milliseconds or in subdivisions of a note by clicking (MS or
BPM).
When you switch from MS to BPM, Serum calculates the nearest subdivision to the millisecond value
for each control based on host tempo.
You can select whether the vertical lines that appear in the graph background are placed at time (ms) or
beat intervals.
Right-click on the graph and select either Time or Beats in the Grid sub-menu.
Modifying Envelopes
You can modify envelope shapes either by setting values using the knobs in the envelope window, or by
manipulating the envelope waveform directly.
You can modify the following envelope parameters using the knobs and directly on the graph:
Control Description
ATK Attack. The time for the initial run-up from start to peak, beginning with the
note on event.
HOLD Hold. The time that the envelope stays at full volume before entering the
decay phase.
DEC Decay. The time for the subsequent run-down from the attack level to the
designated sustain level.
SUS Sustain. The level during the main sequence of the sound’s duration, until
the note off event.
REL Release. The time for the level to decay from the sustain level to zero after
the note off event occurs.
• Attack curve
• Decay curve
• Release curve
Curve Adjustments
Using LFOs
Serum offers ten low frequency oscillators (LFO) that each feature a set of independent controls. The
following shows the LFO area of Serum.
Serum LFOs
Note: When you initialize a patch, only LFO 1 to LFO 6 are visible. LFO 7 to LFO 10 becomes visible
after you use (assign) LFO 6.
Configuring LFOs
Select an LFO by clicking
the corresponding LFO tile.
To copy all LFO settings from one LFO to another, Option-click-drag (macOS) or Alt-click-drag
(Windows) an LFO title to another.
For example, to copy all the LFO 1 settings to LFO 2, Option/Alt-click-drag the LFO 1 tab to
LFO 2.
Note that each tool relies on the current grid setting. The following table describes the LFO drawing
tools available:
Ramp Up Add ramp ups between points, based on the current grid size.
Ramp Down Add ramp downs between points, based on the current grid size.
Modifying LFOs
When modifying points on an LFO graph, you can do the following:
Action Task
The LFO module features a complete set of controls that you can use to set the LFO type, LFO mode,
BPM, grid size, and other options. The following table explains each of the LFO controls:
Control Description
• Normal
• Path
• Chaos: Lorenz
• Chaos: Rossler
• S&H
MODE (Retrig) Specifies how the LFO behaves when a new note is played.
• FREE — The LFO follows the host clock and ignores note timing.
• RETRIG — Retriggers the LFO, causing the LFO to start with a new
note.
Use this setting when you want the LFO to always have the same
timing with new notes.
This causes the LFO to play through, then cycle back to the selected
loopback point.
MONO Select whether the LFO is monophonic or polyphonic.
By default, LFOs are polyphonic allow independent modulation for each
voice. With MONO enabled, the same modulation is applied to all voices.
SHAPE Displays a pop-up menu allowing you to load an LFO preset.
Note that this overwrites the current LFO graph. The menu also offers the
ability to save the current LFO graph as a user-defined preset.
DIRECTION The direction of the LFO, from among the following:
• Forward
• Reverse
• Ping Pong
Control Description
GRID The grid size of the LFO graph. The visual grid background in the LFO graph
changes as you adjust this setting.
Double-click the horizontal or vertical grid box and specify a number to set a
corresponding grid.
The grid is helpful when Alt-clicking (to snap points) or Shift-clicking (to draw
step segments) on the LFO graph. This allows you to align modulation to
the rhythm of your production, or create arpeggiator-like pitch modulations.
HOST Specifies whether the LFO is always synced to the global song position.
When BPM is enabled, the HOST switch determines whether the LFO
playback position “jumps” if you change the LFO rate.
When HOST is enabled, the phase is “anchored” to the host transport
position. For example, when changing the rate from ¼ note to 1 bar, the
phase may jump to have the playback properly fixed to the bar cycle.
BPM/HZ Set the time value to snap to song tempo-based units (1/4 note, 1/8 note,
and so on) or Hertz (free time).
RATE The playback speed of the LFO. This determines the amount of time
represented on the LFO graph area.
The LFO rate is in beat-synced units by default (when BPM is enabled) but
can also be set to a frequency in Hertz (when HZ is enabled).
When BPM is selected (see above), right-click the RATE knob and choose
Swing in the context menu to add swing to the LFO (using the SWING
setting above the keyboard).
When HZ is selected, right-click the RATE knob and choose 10x in the
menu to have the range of the rate control (in Hz) multiplied by a factor of
10, allowing for faster LFO rates.
TRIP/DOT Set triplet and dotted time on the rate control respectively.
These are useful for avoiding triplet or dotted times when automating the
LFO rate, that is, for avoiding dotted and triplet times when you know you
want evenly beat-divisible time.
RISE The amount of time for the LFO graph shape to have influence over the
LFO output.
The LFO begins with a fixed output (imagine the LFO graph as a fat
horizontal line, with the value of the left-most point of the LFOTool graph)
and slowly (based on the rise time) becoming the shape of the visible graph.
This knob is useful for having the LFO slowly influence your sound.
Control Description
DELAY The amount of time before the rise begins. The LFO has a fixed output, as
described above. After the delay time period, the rise begins.
SMOOTH Smooth the LFO output. This is useful for avoiding abrupt jumps in the LFO
output, without having to draw ramps on every segment of the LFO graph.
PHASE Sets the start position of the LFO phase.
Right-click the PHASE knob and choose Snap to Grid in the context menu
to have the phase value snap to the vertical grid lines (as defined by the grid
parameter at the bottom right of the LFO display).
Unlike Serum 1, the HOST switch now has an effect when BPM is disabled.
With BPM disabled, the HOST switch determines whether phase is calculated from the
host transport sample position on retriggering an LFO. One reason to enable the HOST
switch is to ensure that, even though an LFO is set to FREE, it plays back exactly the same
every time a song is played through.
Conversely, a reason to disable the HOST switch is to allow a truly free-running LFO. Each
time playback of a song starts from the beginning, the LFO phase will continue as if it had
been free-running since the last time playback started.
LFO Editor
You can use the same range of tools in the LFO Editor, with the advantage of being able to make finer
adjustments using the larger canvas.
As you are dragging, notice that an LFO 2 label hovers next to the mouse pointer The pointer adds a +
sign as you hover over an assignable knob (in this case, the DETUNE knob).
The + sign indicates that you are over a valid mod destination.
A smaller blue halo appears to the top left of the knob. Hovering over this small halo displays an Up/
Down arrow control.
As you drag the arrow, notice how the halo shrinks or Setting Modulator Depth
expands to show the range of modulation.
Alternatively, as a shortcut, you can also Option/Alt-click and drag on the main knob (the
DETUNE knob itself, for example) to change the modulation depth.
Similar to the smaller blue halo, a gray halo appears to the top left of controls that also have a modulator
assigned, but when the modulator source is not currently selected.
To adjust the modulator depth in this case, simply click the corresponding mod source tab and adjust
the depth using the halo.
Alternatively, you can also adjust the depth using the Modulation
Matrix. See “Using the Modulation Matrix” on page 209 for
complete details.
If you drag to a control that is not centered, such as the filter resonance (the RES knob), Serum assumes
you want the modulation to add value only, and unidirectional is used.
You can change the type setting without visiting the modulation matrix window by Shift-
Option-clicking (macOS) or Shift-Alt-clicking (Windows) on the knob with a visible (blue/
yellow) modulation assignment.
When the value becomes negative, the color of the halo changes to a lighter blue. This indicates that
the depth amount is inverted (as the LFO output goes up, the influence on the control goes down).
To do so, load the appropriate wavetable in the oscillator and use the corresponding WT POS knob to
select the appropriate frame.
Next, select the LFO tab to which you want to copy the wavetable shape, and click the Default menu.
Choose Wavetable A to LFO to copy the currently-selected frame in OSC A to the LFO. The same
applies for oscillators B and C.
LFO point modulation works using LFO busses. Busing allows multiple point modulation by a single
source without requiring you to define duplicate assignments.
• Bus menus
In the LFO display, right-click (Ctrl-click on macOS) a point and choose Modulate X or Modulate Y in
the menu that appears.
Select an LFO bus (initially only LFO Bus 1 appears) and choose a modulation source. The modulation
source you choose will control the LFO point movement. The LFO graph updates to show a shaded bar
representing the modulation range (from minimum to maximum).
You can optionally drag to select multiple points in the LFO graph and configure modulation
for these points in a single assignment.
This allows you to modulate the point either horizontally (X) or vertically (Y), adding the source using the
next available LFO bus.
Mod Source Display all possible modulation sources for the control/knob. Use this
(submenu) menu to quickly configure a connection without having to drag from the
modulation source or visiting the Modulation Matrix window.
Aux Source (submenu) Display all possible auxiliary sources for the control/knob.
Edit Custom Curve Open a curve editor to define a custom mapping applied to the mod source.
Bypass Modulator Bypass the current modulation connection (between the currently-selected
modulation source, such as LFO 1, and the control/knob).
After selecting the option, a check mark appears next to the menu item and
the halo around the control/knob turns gray indicating that the modulation
connection is bypassed. You can reverse this by selecting the menu item
again (uncheck the option).
You can also bypass a modulator by right-clicking the source tile (such as
LFO 1) and choosing the corresponding option.
Remove Modulator Remove the connection between the modulation source and the control/
knob.
Remove All Remove all connections to the control/knob from all modulation sources.
Modulators
Reset Control Reset the control/knob to the default value. This is the same as Cmd-clicking
(macOS) or Ctrl-clicking (Windows) the control.
MIDI Learn Activate MIDI learn mode. When enabled, Serum waits for an incoming
MIDI CC value. After Serum receives a MIDI CC value, MIDI learn mode is
deactivated and the CC# is assigned to the control/knob.
The assignment is saved with the preset (patch).
Remove MIDI cc Remove the MIDI CC# assignment, if any.
Lock Parameter When enabled, lock the parameter (or module) setting (preventing a value
change) when loading presets.
For example, velocity can dynamically influence parameters like volume, filter cutoff, or even the
brightness of the timbre, enabling nuanced expression that responds directly to your playing intensity.
Similarly, you can map note values to parameters such as oscillator pitch, filter resonance, or effects
settings, allowing the sound to evolve based on the pitch being played. With the flexibility to design and
shape the response curve, you can fine-tune how each parameter behaves across the velocity range or
note spectrum.
This approach not only enhances the expressiveness of your sounds but also opens up creative
possibilities for crafting unique, dynamic, and musically responsive presets.
Velocity Settings
You can use the velocity tab to define the MIDI velocity graph that
you can later use to modulate the range of Serum parameters.
Draw the graph using the tools and operations described in the next
section.
Then modulate one or more Serum controls (such as the filter cutoff,
for instance) using the procedure described later in this section.
Velocity Graph
Right-click the VELO tab and enable Legato (Portamento Time) in
the context menu to have portamento applied to the velocity curve
when a note is triggered and another is already held.
Using the same context menu, choose Init Graph to remove all added points and reset the graph to a
straight diagonal line.
The following table describes operations you can perform when editing the graph:
Modulating a Control
Note Settings
You can use the note tab to define the MIDI note graph that you
can later use to modulate Serum parameters.
Draw the graph using the tools and operations described in the
next section.
Using the same context menu, choose Init Graph to remove all added points and reset the graph to a
straight diagonal line.
You can modify the note graph by manipulating the graph directly,
adding new points and dragging curves as needed using your
mouse.
Begin by ensuring that the NOTE tab is selected. Then use your
mouse to adjust the graph.
Modulating a Control
Using Macros
Serum features eight macros that you can use to simplify the control of multiple parameters
simultaneously. Macros provide an efficient way to design, perform, and tweak sounds without having
to manually adjust numerous parameters one by one.
Macros Pane
For example, instead of adjusting filter cutoff, resonance, and pan individually, you can assign these
parameters to a single macro.
Macros are great for experimenting with sound because small changes to a macro setting can result in
complex shifts across multiple sound elements. Macros are also powerful in live performance setups,
where fast and intuitive control is essential.
In addition to being a modulation source, macros can also serve as a destination. This offers
incredible flexibility when setting up modulation.
For example, consider the case where you want a second aux source to modulate a
destination. You could set a macro as an aux source and then modulate it with another entry
in the modulation matrix, using both a main and aux source.
Assigning a Macro
You can assign a macro to a control by dragging and dropping the macro selector to the appropriate
control. As you are dragging, notice that a label hovers next to the mouse pointer The pointer adds a +
sign as you hover over an assignable knob (in this case, the DETUNE knob).
Hover the mouse pointer over the macro to display a small tooltip showing the destinations.
You can repeat this process and assign the same macro to
multiple controls.
Note: Serum offers 64 modulation matrix slots in a patch. You can use these slots to modify or scale up
to 64 destination parameters, one per slot, with 49 different modulation sources.
Serum enhances the typical modulation matrix found on many synthesizers by integrating this matrix
with the drag-and-drop style of routing described in “Exploring Sound Modulation”.
Dragging a modulation source to a control/knob causes the routing to automatically appear in the
matrix and vice-versa. This gives you extra flexibility when viewing, configuring, and modifying your
modulation assignments.
You can use the modulation matrix to configure the following options:
Option Description
Option Description
Option Description
CRV (Curve) Scale the auxiliary source to behave in a non-linear fashion. A setting of 50%
designated linear. The curve is bypassed when it is gray.
OUTPUT Scale the final modulation output, allowing for fine tuning.
Select to bypass this row of the modulation matrix (causing it now to have
no effect).
Click to remove this row of the modulation matrix. This removes the
modulation assignment from the patch.
Click the button (near the top left) to expand the matrix to show more rows. Alternatively, press
Option-F (macOS)/Alt-F (Windows) to expand the view.
Similarly, click the button to revert the modulation matrix down to the default size. Similarly, press
Option-F (macOS)/Alt-F (Windows) to revert the view to the original size.
Moving a Modulation
Bypassing a Modulation
Every row features a bypass button that allows you to easily bypass the modulation.
Clicking the button bypasses the modulation in the signal routing, indicated by the button updating
to show that the row is being bypassed.
Removing a Modulation
You can remove a modulation assignment directly from the modulation matrix. To do so, click the
button for the corresponding modulation.
Click the button (near the top right) and choose an option in the menu that appears.
Operation Description
Sort by Source Sort the modulation matrix (ascending) by the SOURCE column.
Sort by Destination Sort the modulation matrix (ascending) by the DESTINATION column.
Lock Matrix (Keep Lock the modulation matrix, which keeps the modulation assignments when
Assignments on you change presets or initialize a new preset.
Preset Change)
Create Vibrato Create a new modulation that maps the next available LFO to “Main Tuning”
(Unused LFO->Pitch using the Mod Wheel.
via Wheel)
Create Velo->Amp Create a new modulation that maps VELO to the Amp.
Assignment
Apply and Delete “Bake” the macro adjustments into the current preset.
Macros
Specifically, for any parameter assigned to a macro, update the current value
of the parameter to include any offset from the macro. Then remove all
modulation assignments for all macros from the modulation matrix.
Voicing Settings
The VOICING section contains controls that allow you to
change how Serum behaves when multiple notes play at
once.
Mono
Use the MONO switch to enable monophonic mode, causing Serum to only allow one active note at a
time. If a new note is pressed while a note is already playing, the earlier note is interrupted (technically
the note is re-pitched to the new pitch).
Legato
LEGATO is only audible when MONO is enabled. When a monophonic voice is interrupted, the state of
the LEGATO switch determines whether the envelopes/LFOs retrigger.
When LEGATO is enabled with MONO off, Serum behaves paraphonically (especially
noticeable when ENV 2 is modulating a filter). This means you can control envelope
retriggering modes for the effects assignments here also.
When LEGATO is enabled, the envelopes do not retrigger, which results in a smooth change to the new
note. However, sometimes you want the envelopes to retrigger so each note has the same definition; in
these cases, set LEGATO to off.
Note that you can set individual envelopes to behave opposite to the LEGATO setting. See “Inverting
the Legato Setting” on page 185 for more information.
Poly
Use the POLY (polyphony) setting to specify the number of simultaneous notes that can be played.
Sometimes, often for CPU reasons, you might want to place a limit on the simultaneous notes.
For instance if you are sending a flurry of notes (arpeggio) to a patch that has a long release (say 10
seconds), this could end up producing a large number of voices, potentially overloading your CPU.
Typically, eight voices is enough and 16 is generally considered a lot. Note that this control is disabled
when MONO is enabled.
Polyphony Count
This shows the number of voices playing against the total number of voices allowed. In the example,
“0 / 1” indicates that 0 out of a total possible number of 1 voices is playing.
If a note is played, the display would change to “1 / 1”. Similarly, if you enable a second oscillator, the
display would show “0 / 2” when no notes are playing. In other words, Serum totals the number of
voices you have active in a patch.
Single note voice count in Serum includes unison voices and, if one or more oscillators are
set to GRANULAR mode, grain count as well.
Limiting Polyphony
You can choose to limit same note polyphony. This means that if a note (for example, note number 60)
is already sounding, pressing the same note again won’t trigger another layer of the same note. This
prevents “stacking” multiple instances of the same note, ensuring clarity in the sound and avoiding
unintended overlaps or muddiness.
In cases when the number of voices exceeds the polyphony setting, you can specify which active note
should be terminated when a new note is triggered.
Right-click the POLY field and use the the Voice Steal Priority submenu to choose which voices get
priority (are not stolen) when Serum performs voice stealing, from among the following:
Option Description
Newest Terminate the earliest played note.
Oldest Terminate the most recently played note.
Highest Terminate the lowest-pitched note.
Lowest Terminate the highest-pitched note.
Velocity Note with the lowest input velocity.
Portamento Settings
Portamento creates a slow glide/pitch bend from one note
to another. It is most commonly used (and is most useful)
when MONO is also enabled.
In this case, when one note plays and then another, the
pitch slowly changes from the first note to the second.
Porta
Use the PORTA knob to control the rate of glide from one note to another.
Curve
Use the portamento CURVE to adjust the contour of glide from one note to another. If set convex
(typical use), the note pitch departs the beginning pitch quickly and slows down as it nears the
destination note frequency.
If set concave (dragged down below half), the opposite is true. In this case, the pitch slowly departs the
source pitch and later rapidly arrives at the destination pitch.
Always
When ALWAYS is activated, the portamento occurs on a new note even if no note is currently playing.
When disabled, a note must be held for portamento to occur on the (second) note.
Scaled
The SCALED switch is potentially useful for melodic leads when you want a less noticeable portamento
on short intervals.
When activated, the portamento rate is adjusted based on the distance between the source and
destination pitches. For example, if the portamento is a glide between two notes one octave apart, the
PORTA knob time value is used.
If the portamento is a glide between notes less than one octave apart, the time is faster and vice-versa
(notes larger than an octave are progressively slower).
A MIDI clip contains notes and controller data for playing your current Serum patch. Each clip specifies
the note pitch, length, position, and dynamics (velocity).
Clips Interface
Use the GLOBAL pane to load a clip bank preset. Alternatively, if you already have clip slots already
populated, you can initialize the CLIP module to start over.
Use the CLIP SETTINGS pane to set clip parameters, including the clip length, transposition, mode,
rate, launch quantization, and more.
Optionally set the key and scale to have Serum quantize the pitch of notes generated by the CLIP or
ARP module, or from MIDI input..
Populate the piano roll with appropriate notes. You can “click in” the notes or record a live
performance.
Optionally, assign macros to a certain clip settings, allowing you to change one or more sound
parameters using a single knob.
Specify whether Serum should output the MIDI data it generates internally.
Bank Menu
Use the TRIGGER MODE switch to set the CLIP mode to MONO or POLY. When
set to MONO, only a single clip can play at one time. When set to POLY, you can trigger multiple clips
to play simultaneously.
Enable the EDIT ALL switch to have your parameter edits apply to all clips. When disabled, any
changes to the clip settings (such as TRANSPOSE or RATE) apply only to the currently-selected clip.
When enabled, changes to any parameter become immediately effective for all clips.
Hold the Option (macOS) or Alt (Windows) key when editing a clip parameter to apply the
change to all clips.
Bank Loaded
Click the BANK field and choose Save Clip Bank in the menu that appears.
If you would instead prefer to overwrite the existing clip bank, you can choose the existing file name in
the dialog.
Piano Roll
Clip Slots
Double-click to add a new note. The latest note you added appears in orange.
Important: When pasting a time range, unlike pasting notes, all existing notes in the destination range
are overwritten.
Moving a Note
Arrow Description
You can perform a series of other operations on notes, as described in the following table. In all cases,
select the appropriate notes and right-click to choose the operation. In addition, in most cases, there’s a
keyboard shortcut to go with the operation.
Conform to Cmd-K/Ctrl-K Move the selected notes to the selected scale (if necessary).
Scale You can select the scale in the Keyboard pane.
Legato Cmd-L/Ctrl-L Apply legato to the selected notes, extending each note to
smoothly connect to the start of the next.
Mute 0 (zero) Mute the selected notes. The muted notes appear in gray.
Quantize Cmd-Q/Ctrl-Q Quantize the selected notes to the current grid setting.
Reverse Cmd-R/Ctrl-R Reverse the order of selected notes.
Scale Time / Halve the time and duration of the selected notes without
50% changing the clip length.
Scale Time * Double the time and duration of the selected notes without
200% changing the clip length.
Double Entire Cmd-E/Ctrl-E Append a duplicate of the clip, copying the notes and
clip doubling the clip length.
Scale Entire (no keyboard Halve the clip length and halve each note’s duration.
clip 50% shortcut)
For example, if the clip is currently two bars long with 8th
notes, this decreases the clip to one bar and halves the note
durations to 16th notes.
Scale Entire (no keyboard Double the clip length and double each note’s duration.
clip 200% shortcut)
For example, if the clip is currently one bar long with 16th
notes, this increases the clip to two bars and doubles the
note durations to 8th notes.
Select All Cmd-A/Ctrl-A Select all notes in the clip.
To delete a single note, double-click on the note. To delete multiple notes, select the notes and press
the Delete key (macOS) or Backspace key (Windows).
Normally, when you select a note, no sound is played. You can have the note sound when selected by
toggling the headphones on. The button changes to show that sound is enabled.
Note that key commands on the piano roll only work if the note grid or automation lane has keyboard
focus. You can give either focus by clicking on the corresponding background.
A keyboard icon appears near the top right to show which pane has keyboard focus.
The Keyboard shortcuts preference needs to be set to ON in the GLOBAL pane to enable key
commands to work.
The grid updates as you drag the control. This gives you incredible creative flexibility in terms of rhythm
and precision. A smaller grid size allows for more precise note placement, which is useful when fine-
tuning timing or adding nuanced variations.
For example, you could use a larger grid size (such as quarter notes and eighth notes) to create standard
rhythms like 4/4 beats or simple melodic lines. Smaller grid sizes (such as 16th notes, 32nd notes, or
even smaller) enable more detailed rhythmic work, such as rapid hi-hat patterns, rolls, or syncopated
melodies.
Grid Menu
• Hold Cmd-Option (macOS) or Ctrl-Alt (Windows) and drag the piano roll background to scroll in
any direction.
• Hold Shift-Option or Shift-Alt and drag the piano roll background to zoom in any direction.
• Hold Shift-Cmd or Shift-Ctrl and drag on the note grid to drag a (black) marquee. The grid zooms
to the selected area after releasing the mouse button.
• Use the mouse wheel to scroll the piano roll vertically or, with Shift held, horizontally.
• Hold the Option or Alt key and use the mouse wheel to zoom the piano roll vertically. Hold the
Cmd or Ctrl key with the mouse wheel to zoom horizontally.
You can also use the mouse wheel while hovering over the timeline (near the top) to zoom
horizontally.
You can “fold” the notes on the piano roll to hide unused notes, making the piano roll appear cleaner
and less cluttered.
The fold feature becomes especially useful when a key and scale are selected. In this case, the piano roll
collapses to display only the notes within the scale.
Hold the Cmd (macOS) or Ctrl (Windows) key when clicking the button to show notes
in the scale as well as used notes.
This makes it easy to distinguish in-scale notes (blue background) from out-of-scale notes (gray
background).
For example, the following shows notes in the C Major scale when the C Minor scale is selected and the
piano roll is folded.
Option Description
Wrap Clip to Reset the start offset marker to the same position as the start marker
Begin Here moving all notes and automation with it.
Any events that came before the start offset marker are wrapped around to
the end of the clip.
Quantize Quantize the position of the start offset marker when modulation is applied
to the marker.
Also, determine the time division by which each successive key will offset
the playback start position when KB Span is set to Offset (see Configuring
Clip Settings).
CLIP Field
Clip Settings
The following table describes the remaining parameters you can set:
Field Description
• Mono and Poly — Play the clip transposed according to the played note
(relative to C3).
• Offset — Play the clip starting from a different position for each note,
determined by the offset quantize setting (accessed by right-clicking
the start offset marker).
• Off — Turn the setting off.
TRANS The transpose amount for the clip.
MODE The play mode for the clip, specifying how the playhead moves. Note the
following about the available options:
Note: If you enabled the EDIT ALL switch in the GLOBAL pane, any changes you make to the
clip settings apply to all clips. When disabled, any changes to the clip settings apply only to the
currently-selected clip.
Managing Clips
Serum features 12 clips per clip bank. As you’ll see in the next section, this makes it particularly easy
and convenient to trigger these clips using either a MIDI keyboard or even your computer keyboard.
Clip Slots
Click a clip to select it. The current clip is highlighted, and the contents of the clip appear in the piano
roll.
The following table describes the available operations: Current Clip (Highlighted)
Operation Description
Rename Clip By default, clips are named 1 to 12 (from left to right). You can rename any
clip to better match the clip contents.
After choosing Rename Clip in the menu, a text area appears in the clip
header.
Type the new clip name and press the Enter key.
Copy Clip Copy the clip contents (notes) and settings (including the clip name) to
Serum’s internal clipboard.
Right-click the source clip and choose Copy Clip in the menu. Right-click the
destination clip and choose Paste Clip in the menu.
Note that it is not possible to copy a clip from one instance of Serum to
another.
Paste Clip Paste the most-recently copied clip into the target clip.
Erase Clip Erase the contents of the selected clip.
Set as Preview Clip Set the clip as the MIDI sequence to play when someone previews the
saved preset in the Serum presets browser.
When designing sounds, both for yourself and others, it’s helpful to set a preview clip. This
gives preset users a quick sense of the intent of the preset. Otherwise, Serum plays a
designated “fallback” clip, which can quickly get repetitive when previewing a lot of presets.
Plus, by including one or more clips with your preset, you offer yourself and others a bonus
when you choose to use the sound in the future!
Triggering Clips
You can trigger your clips in multiple ways, allowing clips to be a versatile part of your sound design
process.
You can make changes to any aspect of the clip while it is playing, including modifying notes, changing
clip settings, and more. You can also change the underlying sound design as the clip plays, providing a
versatile way to hear your sound evolve in context.
This is helpful when you are not in CLIP mode and instead
tweaking the oscillators, mixing, applying effects, or adjusting
the modulation matrix.
Notice that the clip slots (in CLIP mode) are arranged similar to Clip Keyboard
a keyboard. This makes it easy to logically map the clips slots
to the CLIP keyboard.
If you launch a clip without notes, the key highlights in green but the rhythm indicator is not present.
Press the same key to stop the clip, or press another key to launch the corresponding clip.
Important: If you launch a clip that you know has notes but don’t hear any sound, verify that clip
playback is enabled.
OVERDUB loops continuously using the start and end markers that you set, allowing you to add new
notes during each pass. EXTEND, in contrast, extends the piano roll timeline until you either stop
recording or switch to OVERDUB mode. You can change between the two modes at any time.
Click the button to commit your most recent performance to the clip. The notes change from red
to green to indicate the committed status.
After committing, any new notes that you play appear in red (while the committed notes remain green).
Click the button to undo your last performance (notes not committed to the clip).
When you’ve finished, click the button. All notes become “committed” and appear in green.
Note, however, that the clip continues to play.
To stop playback, click the stop button for the corresponding clip.
The specific setup to do this depends on your DAW. The following describes how to configure this
using Apple Logic Pro X.
Important: Remember that MIDI OUT in Serum needs to be enabled (set to either On or Clip Player).
CLIP Menu
Operation Description
Lock Module Enable this option to lock the entire CLIPS module, including all loaded clips,
when changing (sound) presets.
MIDI Input Trigger Select the octave from which notes trigger clips. Alternatively, set to Off if no
Octave triggering is required.
By default, this is set to the lowest octave of the MIDI note range.
Using Macros
You can assign macros to a number of controls within the CLIP module, thereby allowing you to change
one or more sound parameters using a single knob.
In addition, you can use any existing macro assignments to tweak the sound without having
to leave CLIP view. Furthermore, you can record macro automation to a clip by dragging the
macro knobs.
To assign a macro to a control, drag and drop the macro selector to the appropriate control. The +
sign indicates that you are over a valid modulation destination. Note that not all controls are valid
destinations.
When you release the mouse button, Serum automatically makes the connection causing the macro to
now affect the destination control.
After setting the macro, notice that a number 1 now appears next to MACRO 1. This indicates that
MACRO 1 has one destination.
Hover the mouse pointer over the macro to display a small tooltip showing the destinations.
You can repeat this process and assign the same macro to multiple controls. You can then manipulate
the macro (perhaps assigned to a knob or slider on a physical controller) as you would a mod wheel.
Serum offers a broad array of patterns, fully customizable through parameters such as transposition,
offsets, repeats, and chance, among others, giving you precise creative freedom.
Using the arpeggiator, you can swiftly craft intricate, evolving musical expressions that transform simple
chords into captivating, complex melodies.
Serum Arpeggiator
Arpeggiator Module
Use the GLOBAL pane to set parameters that affect all aspects of the arpeggiator, including the launch
quantization and whether parameter edits affect all arp slots. You can also use the GLOBAL pane to
load arpeggiator presets, create a new arp bank, and save your own arp bank as a user-defined preset.
Use the PATTERN pane to set the shape of the currently-selected arp slot (or all arp slots if you
selected that EDIT ALL option in the GLOBAL pane). You can also use this pane to set the arpeggiator
rate (in either beats per minute or Hertz), and enable triplets and dotted notes.
Use the TRANSPOSE pane to set the transpose shift (positive or negative) and range for the
arpeggiator (which specifies how many times the pattern is transposed by the shift setting). You can also
set the shape of the transpose range, which further expands your creative possibilities.
Use the PLAYBACK pane to specify the playback settings for the arpeggiator, including offsets and
repeats, as well as the length of the gate and the probability of a note being played. You can also enable
a latch to continue playing notes without needing to keep the keys pressed down, as well as pass
incoming notes to the output (similar to a MIDI THRU port on a device) .
Use the RETRIGGER pane to specify how the arpeggiator retriggers the arp shape/pattern. You can set
to retrigger the arp when the slot is launched or on a incoming note (first note or otherwise). You can
also specify the rate at which the arpeggiator is retriggered.
Use the VELOCITY pane to configure the arpeggiator to raise or lower the arpeggiator note output
velocities over time. You can set the speed at which the velocities are changed (decay) as well as the
target velocity towards which the decay setting moves.
The ARP module further features 12 arpeggiators per arp bank. This offers a rich playground in which to
explore and define evolving rhythmic patterns and dynamic musical phrases.
Bank Menu
Enable the EDIT ALL switch to have your parameter edits apply to all arp slots.
When disabled, any changes to the arp settings (such as TRANSPOSE or CHANCE) apply only to the
currently-selected arpeggiator slot.
When enabled, changes to any parameter become immediately effective for all arp slots.
Hold the Option (macOS) or Alt (Windows) key when editing an arp parameter to apply the
change to all arps.
If you would instead prefer to overwrite the existing arp bank, you can choose the existing file name in
the dialog.
Click the TRIP button to enable triplets. Click the DOT button to enable dotted notes.
Important: The parameters you set in this and the other panes only affect the currently-selected arp
slot unless you enabled EDIT ALL in the GLOBAL pane..
One way to try all the various shapes is to enable the LATCH button (in the PLAYBACK
pane), play a chord on your MIDI controller, and then cycle through all the shapes one by one
to hear the difference. You can also watch the Serum keyboard to get a visual sense of how
the shape arpeggiates the chord.
Pattern Option
Use the left panel to specify the pattern settings and the right panel to configure the arpeggiator graph.
Arp Editor
Loading a Pattern
You can quickly load a factory-supplied
or user-defined pattern.
For example, you can use the pane to set the pattern
length, the mode (normal, reverse, and so on), the step
mode, and the wrap settings.
Pattern Option
Field Description
• Normal
• Reverse
• Pendulum
• Random
• Rand Start
• Rand End
• One Shot
• Static
TIME For random play modes, specifies how often the playhead jumps to a new
random position. You can choose from values between 1/16th note and four
bars.
STEP MODE Specifies how each step is played, from among the following:
• Normal
• New Only
• Chord
• Chord (new)
The chord triggers all held notes on each step with voicing determined by
step numbers.
Note that new modes trigger a step only if the input note is received exactly
at the same time as the step triggers.
WRAP Specifies how pattern steps less than, or greater than, the number of held
keys are treated.
PITCH The wrap transpose, from 0 to 24 semitones.
RANGE Specifies how pattern steps less than, or great than, the number of held keys
are transposed.
Saving a Pattern
After creating a new pattern, or modifying an
existing pattern, you can save the configuration
as a new pattern.
Click the pattern name (topmost) field and choose Save Pattern in the menu that appears.
Renaming a Pattern
You can rename a pattern, as needed.
Renaming a Pattern
Graph Editor
Click the grid size and drag up or down to change the default grid setting.
Use the automation lanes to control various parameters over time. Right-click in various places in
the arp grid to display a context menu for the corresponding element.
Transpose Settings
You can set the transpose shift and range for the
arpeggiator.
Click and drag the SHIFT knob to set the amount that
each repetition of the pattern is transposed. This can
be a positive or negative value.
The following table describes the range shape options available. Note that each example uses a SHIFT
setting of 3 and a RANGE setting of 4.
Shape Description
Up Transpose the arpeggiated notes by the SHIFT setting and repeat the
number of times specified by the RANGE setting.
Down Transpose the arpeggiated notes by the SHIFT setting and repeat the
number of times specified by the RANGE setting.
Up/Down Transpose the arpeggiated notes by the SHIFT setting and repeat the
number of times as specified by the RANGE setting.
Then immediately reverse (transpose the arpeggiated notes in the reverse
direction and repeat using the same settings).
Shape Description
Down/Up Transpose the arpeggiated notes by the SHIFT setting and repeat the
number of times specified by the RANGE setting.
Then immediately reverse (transpose the arpeggiated notes in the reverse
direction and repeat using the same settings).
Up+Down Transpose the arpeggiated notes by the SHIFT setting and repeat the
number of times specified by the RANGE setting.
Then start again in the reverse direction.
Down+Up Transpose the arpeggiated notes by the SHIFT setting and repeat the
number of times specified by the RANGE setting.
Then start again in the reverse direction.
Shape Description
Shape Description
Pinky Up Transpose the arpeggiated notes to the maximum (as specified by the
SHIFT and RANGE settings). After each transposition, play the maximum
transposition followed by the original arpeggio transposed.
Repeat the number of times specified by the RANGE setting.
Pinky UD Transpose the arpeggiated notes to the maximum (as specified by the
SHIFT and RANGE settings). After each transposition, play the maximum
transposition followed by the original arpeggio transposed.
Repeat the number of times specified by the RANGE setting.
Then immediately reverse (transpose the arpeggiated notes in the reverse
direction and repeat using the same settings).
Shape Description
Converge Transpose the arpeggiated notes to the maximum (as specified by the SHIFT
and RANGE settings). After each transposition, converge the transpositions
between original to maximum.
Diverge Transpose the arpeggiated notes as specified by the SHIFT setting. After
each transposition, diverge the transpositions further.
Con+Diverge Transpose the arpeggiated notes first using the converge pattern followed
by the diverge pattern (see above).
Shape Description
Chord Use the SHIFT and RANGE settings to play the chord instead of the
arpeggio.
Random Randomly transpose the arpeggiated notes based on the SHIFT setting and
repeating the number of times specified by the RANGE setting.
Shape Description
Rnd.Drift Randomly transpose (with a bias towards drifting) the arpeggiated notes
based on the SHIFT setting and repeating the number of times specified by
the RANGE setting.
Rnd.Once Randomly transpose the arpeggiated notes once based on the SHIFT setting
and repeating the number of times specified by the RANGE setting. Then
continue using this pattern.
Playback Settings
You can specify the playback settings for the arpeggiator, including offsets and repeats, as well as the
length of the gate and the probability of a note being played.
Click and drag the following knobs to perform the associated action:
Knob Description
When the arp is enabled, MIDI CC64 sustain messages sent to Serum will control the latch
instead of regular note sustain.
Retrigger Settings
You can specify how the arpeggiator retriggers the arp shape/pattern.
Enable the LAUNCH setting to retrigger the arp when the slot is
launched.
Enable the RATE setting and set an appropriate rate by clicking and
dragging the field. The rate sets the interval at which the arpeggiator is
retriggered.
Velocity Settings
You can set the arpeggiator to raise or lower the arpeggiator note
output velocities over time. Enable the VELOCITY setting to activate
the velocity settings (in this section).
Enable the RETRIG setting to reset the velocity decay value (back to
incoming velocities) whenever the RETRIGGER feature is activated.
(See the previous section for more information about retrigger settings.)
Click and drag the DECAY knob to set the speed at which the velocities
are changed.
Click and drag the TARGET knob to set the target velocity towards which Velocity Settings
the DECAY setting moves.
Managing Arpeggiators
The ARP module features 12 arpeggiators per arp bank (similar to the CLIPS module). This makes it easy
and convenient to trigger these arps using either a MIDI keyboard or your computer keyboard.
Arpeggiator Slots
Click an arp slot to select it. The current arpeggiator is highlighted and the play button turns purple. The
arp settings appear in the various sections.
Operation Description
Rename Pattern By default, arpeggiators are named after the pattern shape, such as Up (1/16)
or Down (1/32).
When you define the pattern yourself (by choosing Pattern in the SHAPE
field and then using the pattern editor), you can rename the arp to better
describe the pattern you created.
Right-click in the arp and choose Rename Pattern in the context menu. A
text area appears in the clip header.
Type the new pattern name and press the Enter key.
Copy Pattern Copy the arp settings to Serum’s internal clipboard.
Right-click the source arp and choose Copy Pattern in the context menu.
Right-click the destination arp and choose Paste Pattern in the menu.
Note that it is not possible to copy a pattern from one instance of Serum to
another.
Paste Pattern Paste the most-recently copied arp into the target arp.
Erase Pattern Erase the contents of the selected arp.
Show Pattern Names Show the arp names in the slots.
You can make changes to any aspect of the arp while it is playing, including changing patterns,
modifying arp settings, and more. You can also change the underlying sound design as the arp plays,
providing a versatile way to hear your sound evolve, in context.
This is helpful when you are not in ARP mode and instead
tweaking the oscillators, mixing, applying effects, or adjusting
the modulation matrix.
Notice that the arp slots (in ARP mode) are arranged similar Arpeggiator Keyboard
to a keyboard. This makes it easy to logically map the arp
slots to the ARP keyboard. Press a key to play the corresponding arp. The key highlights in purple.
Press the same key to relaunch the arp, or press another key to launch that arp.
Important: If you launch an arp but don’t hear the sequence that you expect, verify that arp playback is
enabled.
ARP Menu
Operation Description
Lock Module Enable this option to lock the entire ARP module, including all arps, when
changing (sound) presets.
MIDI Input Trigger Select the octave from which notes trigger arps. Alternatively, set to Off if no
Octave triggering is required.
By default, this is set to the lowest octave of the MIDI note range.
Serum Keyboard
Use the TRANSPOSE field to transpose the keyboard within a range of two octaves (set in
semitones).
Use the KEY and SCALE fields to choose appropriate options. Serum offers a very thorough list of
scale options from which to choose.
Use the SWING field to set the swing applied to certain notes to add groove.
And finally, use the OSC MAPPING feature to edit the note ranges of the oscillators and
arpeggiator.
Transpose
You can transpose the keyboard within a range of -24 to +24 semitones.
Keyboard Transpose
Click the TRANSPOSE field and drag up or down. Alternatively, for even finer control, click the up and
down arrows to adjust the setting by a single semitone.
You can also hover over the field and adjust the setting by 12 semitone increments using the mouse
wheel.
Scale Options
After selecting the key and scale, the piano roll in the CLIP module highlights the root note of the key.
In addition, notes within the scale are highlighted with a gray background, while notes outside the scale
appear with a darker gray background.
Setting Swing
You can set the swing applied to certain notes to add groove, thereby imparting a more human feel
instead of a rigid, machine-like rhythm.
Swing
Click the SWING field and drag up or down. The range of swing values depends on the host DAW that
you are using (see the note below). You can also hover over the field and adjust the setting using the
mouse wheel.
Click and drag to set the appropriate division. You can also use a mouse wheel or the arrow buttons.
Serum attempts to match the swing value displayed to the convention used in the host
DAW. This means that setting the same values in Serum and the host should result in the
same amount of swing.
For example, when using Ableton Live, swing is displayed as a range between 12.5% and
87.5%. In contrast, FL Studio displays swing with a range of -150% to 150%.
Oscillator Mapping
You can edit the note ranges of the oscillators and arpeggiator to define and limit the range of notes to
which the oscillator or arpeggiator responds.
Oscillator Mapping
Click the OSC MAPPING button. The oscillator mapping dialog appears.
Click and drag the left or right edge of the note range for
the corresponding oscillator or arpeggiator.
Key Mapping
The current note appears below the cursor, and a highlight
shows the note in the context of the keyboard.
The current note boundaries appear below the cursor (hand), and highlights show the note range in the
context of the keyboard.
Use the FOLD checkbox to specify what happens to notes outside the selected range. When
unselected, the notes are ignored. When selected, the notes get “folded” into the selected range. You
can use this, for example, to restrict the range of an oscillator used for sub-bass to a single low octave,
regardless of the octave the incoming note was played.
Select the WARP checkbox to set the corresponding oscillator to warp other oscillators when keys are
played outside of its note range. For example, consider the case when the OSC A warp is set to FM
from B. If you want the warp to apply regardless of the OSC B note ranges, select the WARP checkbox
for OSC B. This allows you to restrict the OSC B note range without losing the warp on OSC A for
notes outside of the OSC B range.
You can assign different oscillators to specific note ranges to create multi-layered sounds or
keyboard splits. For example, you might create bass and lead layers, assigning one oscillator to low
notes for a bass sound and another to high notes for a lead.
Similarly, you could create complex pads by layering different oscillators with unique timbres across
the keyboard range for rich, evolving textures.
Many acoustic instruments have distinct tonal characteristics across their range. Assigning
oscillators to specific note ranges can mimic this behavior.
For example, you could consider simulating how low strings sound warmer while higher strings
sound brighter on a string instrument.
You could create timbral variation by assigning unique waveforms and effects to different note
ranges. For example, you could consider using a sine wave oscillator in the lower range for pure
bass tones and a saw wave in higher ranges for harmonic richness.
This could thereby enhance expressiveness by varying the oscillator’s contribution based on the
note range.
Click and drag the left or right edge of the velocity range for the corresponding oscillator or arpeggiator.
The current velocity appears below the cursor, and a highlight shows the velocity in the context of the
complete range (1 to 127).
The current velocity boundaries appear below the cursor (hand), and highlights show the velocity range
in the context of the complete range (1 to 127).
Wavetable Editor
Thumbnail overviews provide an easy way to view, select, and reorder the subtables that make up a
wavetable.
Flat Line The default tool. Draw a flat line in the grid step.
Slope Down Draw a diagonal line downwards. This is useful for drawing
small saws to give your waveform a buzz.
Interpolate linear Rather than drawing, this tool connects endpoints to draw
a straight line across the grid segment.
This can be useful for smoothing out a waveform (LPF).
Interpolate curved Similar to interpolate linear, however with a more gradual
transition (not as much harmonics as linear; even more
low pass/smoothed).
Nudge Move a portion of the waveform up or down. Use this for
clipping (if you click-drag vertically far enough, the audio
begins to clip/flat-line).
Noise Add noise to the waveform. Dragging vertically up adds
noise, while dragging down reduces noise.
FFT Area
Use the FFT area to draw the waveform harmonics and their relative phases to one another.
The top section shows the frequency bins. The left-most bin represents the fundamental or single sine
wave at the oscillator’s base frequency (the note you play).
The vertical bars to the right represent the harmonics that make up the sound, (2:1 is the octave, 3:1 is
+19 semitones, 4:1 equals 24 semitones, and so on).
The bottom section shows the phase offset for each harmonic. Right-click to display the context menu.
Randomize Low x Bins Insert random values for the left-most number of bins, but allows 50%
(with Half) amplitude for harmonics. This provides more tonal options.
Randomize All Randomize all bins. This results in a noisy/buzzy tone. This is useful as
a starting point for further destruction with FFT drawing or other menu
commands.
Create Random Series Randomly remove harmonics.
Gaps
Progressive Fade Reduce the high frequency content gradually. This is useful for bringing back
a natural-sounding tonal balance to an overly bright or buzzy wavetable
without resorting to removing the highs altogether.
Shift Octave Up Spread all bins (1->2, 2->4, and so on) so that the resultant waveform is
doubled in frequency.
Shift Octave Down Same as above, but in reverse.
Repeat Bin Group Create copies of the harmonics from the start to the current cursor location,
(Start to Cursor Pos) repeating all the way through the spectrum.
Draw Even Harmonics Prevent the mouse from operating on odd-numbered bins.
Only
Draw Odd Harmonics Prevent the mouse from operating on even-numbered bins.
Only
Snap Vertical Draw to Snap vertical mouse drawing to exactly 0%, 25%, 50%, 75% and 100%.
Quarters
Scale Freq Values by Amplitude draw/react proportionally to the selected bin, exaggerating higher
Bin Index harmonics visually. Using this feature, a proper sawtooth will be a horizontal
line draw instead of an exponential curve across the harmonics.
Start by selecting the frames you want to copy. You can click to select a frame, and Shift-click to extend
your selection. Your selection needs to be contiguous.
Select the starting frame of the destination and click the PASTE button.
Click the button and choose an option using the menu that appears. The following table
describes the menu options:
Init All (Default) Clear all frames/interpolations. The wavetable is brought to its default state
(a single saw frame).
Insert (at current Add an additional frame to the wavetable set, inserted after the currently-
index) selected frame. This has the same effect as pressing the + button below the
waveform.
Remove (current Remove the currently selected frame. This is the same function as pressing
index) the - button (below the waveform).
Remove: Remove multi-selected frames (Shift-click in the thumbnail area to multi-
multiselection select a range of frames).
Remove: beginning-> Remove the frames from the beginning (1) through the selected frame.
selected
Remove: selected-> Remove the frames from the selected frame through to the end.
end
Remove All (except A “crop” feature, useful when you decide that a single frame or certain range
Selected) (multiselect) is all you want to keep.
Reduce to Thins the number of frames; helpful to easily keep only every nth frame.
Sorting Frames
You can use the SORT menu to reorder existing frames based on a spectral property. Keep in mind that
you can also manually sort frames by dragging the frames left and right within the thumbnail display.
Click the button and choose an option using the menu that appears. The following table
describes the menu options:
Sort by spectrum Sort frames based on which frame has the highest peak frequency bin.
(Peak Spect)
Sort by spectrum Sort frames based on where the average spectral content exists (sum of all
(Average Spect) frequencies).
Sort by spectrum Sort frames by the highest overall peak (concentrated frequency energy).
(Peak Amount)
Sort by spectrum ( Sort frames by how many frequency bins contain energy (spectral
Num w/Spect) complexity).
Sort by spectrum Sort frames by the highest frequency bin to contain spectra. This should
(Highest w/Spect) work well for filter sweeps, for instance.
Sort by spectrum Sort frames by the amount of energy in the fundamental. When in doubt, try
(Fundamental Amt.) this sort first.
Sort Randomize Randomize the order of the table indices.
Reverse Entire Table Reverse all frames in the table. This is useful if you decide that you want the
Order table order to become bright-to-dull instead of dull-to-bright, for instance.
Importing a Wavetable
You can import a wavetable into the Wavetable Editor.
Click the button, and choose an import option in the menu that appears.
A dialog appears allowing you to select a file to load (which overwrites the wavetable set you are
currently editing). Refer to the “Importing Audio” chapter for a detailed description of the options in this
menu.
Keep in mind that you can also drag-and-drop waveforms directly from the macOS Finder or Windows
Explorer. This is often a more convenient way to import sounds.
Exporting a Wavetable
You can export a wavetable from the Wavetable Editor.
Click the button, and choose an export option in the menu that appears.
Export All As 8-bit Export the wavetable in 8-bit format. This saves the wavetable using a file
(.256) format compatible with certain hardware modular synthesizers (such as
Wiard and Piston Honda).
Export All As 16-bit Export the wavetable in 16-bit format. This saves the file in WAV format,
(.wav) mono, 16-bit, 44100 Hz, 2048 samples per subtable.
Note that additional header information for interpolation mode and
interpolation tables are not saved as part of the WAV data.
Export All As 32-bit Export the wavetable in 32-bit format. This saves the file in WAV file format,
(.wav) mono, 32-bit, 44100 Hz, 2048 samples per subtable.
Note that additional header information for interpolation mode and
interpolation tables are not saved as part of the WAV data.
Export All As Single- Export the wavetable as a set of individual single-cycle WAV files.
Cycle Waves
Export Selection Export and save a wavetable using the Serum file format (a WAV file with
extra header information). This is similar to the other save commands, except
that only the selected frames are included in the file.
If you don’t see a format compatible with your needs, you can convert WAV files to an alternate format
using a third party application or utility.
Single Menu
The SINGLE menu contains process functions that alter the currently selected (visible) frames.
All Menu
The ALL menu offers process functions, similar to the SINGLE menu, but applying to all frames (1-256)
in the wavetable.
Normalize Each Individually normalize every frame to its own peak maximum.
(Gained Separately)
Normalize Same (Max The entire wavetable set is scanned for a peak level, and the same gain
From All Frames) amount is applied to each frame.
Remove DC Offset Remove any DC (up/down bias to the waveform). Noet that this is typically
not needed.
Flip Vertical (Polarity Perform a polarity invert operation (often somewhat mistakenly called
Invert) “phase invert”).
Flip Horizontal Reverse the audio. As a short looping wave, it probably won’t sound
(Reverse) reversed in most situations.
Fade Edges (Grid Size) Create a fade in at the left edge and a fade out at the right edge based on
the width of the horizontal grid size (set in the lower right).
X-Fade Edges (Grid Similar to above, but instead of fading to the center line, the edges fade to
Size) each other.
X-Fade Edges (16 Same as above, but only a micro-fade on the edges (16 samples is small!).
Samples)
Filter (Grid Size) Completely remove upper harmonics using a low pass filter. A lower grid
size means more low pass filtering. You can achieve the same low pass filter
effect using the FFT area (Ctrl/right-click > Clear HF).
Since this uses FFT, it is infinitely steep; you might notice some DC shift as
all DC offset is removed.
Remove Fundamental Remove the lowest frequency (pitch), effectively akin to zeroing the top-left
(HPF) bar of the FFT display (fundamental) but for all tables.
The formula “z=(q>1)?in:0” would also yield the same result, or simply
“(q>1)?in:0” if you want to only apply to the currently visible frame.
Remove Low Spectra Remove the lowest frequency bins, somewhat similar to a high-pass-filter.
(Grid Size)
Remove Low Phases Zero the phases on the lowest frequency bins.
(Grid Size)
Sample Redux at Grid Perform a sample rate reduction for a lo-fi sound.
Size
The grid size is not used in a literal way but as a means for specifying an
amount (larger number offers more sample redux).
Resize Tables to be Divide every frame in half to become two frames. This doubles the number
Half (2x Total of frames (so you shouldn’t use this if you have more than 128 frames).
Number)
Resize Tables to be Every frame is grouped with the following frame. This is useful if you notice
Double (½ Total every second frame similar after a WAV import (the frequency was detected
Number) an octave too high).
Create PWM from Remove all frames except the current frame and create a PWM shift across
This Table to All all 256 frames.
There is a real time PWM effect using the WARP knob (main panel), but this
way you can perform pulse width modulation and use another warp effect.
Nudge All Phases for Preserve the phase of the lowest frequency (fundamental) across various
Fundamental to 50% waveforms.
Set Spectra (This Apply FFT bins (spectrum) from the current frame to all frames.
Frame to All)
Set Phases (This Apply the FFT phase information from the current frame to all frames. This
Frame to All) is useful to make all frames match in phases beyond just the fundamentals,
for example, solid/consistent sound during morphing.
Set Spectra from Remove the relative amount of harmonic content contained in the other
Other Osc oscillator.
Set Phases from Apply the FFT phase information from the other oscillator frames (A<>B) to
Other Osc the frames of the visible oscillator.
Subtract Spectra from Subtract harmonic content based on the harmonic content of the other
Other Osc oscillator’s wavetable.
Blur Spectra - Interpolate (smooth) the harmonic content between adjacent harmonics. The
Adjacent Bins grid size value determines how many neighboring harmonics are factored
(Grid Size) into the smooth operation.
Blur Phases - Interpolate (smooth) the phase content between adjacent harmonics. The
Adjacent Bins grid size value determines how many neighboring harmonics are factored
(Grid Size) into the smooth operation.
Blur Spectra - Interpolate (smooth) the frequency content between adjacent frames. The
Adjacent Frames grid size value determines how many neighboring harmonics are factored
(Grid Size) into the smooth operation.
Blur Phases - Interpolate (smooth) the phase content between adjacent frames. The grid
Adjacent Frames size value determines how many neighboring harmonics are factored into
(Grid Size) the smooth operation.
Shift Horizontal to Move the audio data to the right to have the left edge of the frame begin
Zero-Crossing at a zero crossing (where the waveform polarity changes from negative to
positive).
Morph Menu
Use the MORPH menu to create or remove interpolation frames between existing frames. Note that
you must have more than 1 and less than 256 frames.
After selecting a morph item from the menu, you’ll notice that there are now 256 frames in the WT
POS selector. This is because all empty wavetable slots are now filled with interpolated (blended)
waveforms of the neighboring tables.
You’ll also notice that the thumbnails no longer display “1, 2, 3”, and so on, but rather “1, 17, 33” (or
similar). This is because the intermediate tables (for example, 2-16 in this example) contain the newly-
created morph tables.
You’ll see these intermediate tables in the waveform area as WT POS is adjusted, but they don’t exist as
thumbnails.
After you exit the Wavetable Editor and return to the main panel, you’ll also notice the interpolated
frames appear gray in the 3D wave overview (whereas green shows the “real” frames, and yellow shows
the currently-selected frame).
Morph - Crossfade Create interpolated frames by crossfading the neighboring frames together.
This is the recommended default, and what traditional wavetable synths do.
Morph - Spectral Use the spectral and phase content of neighboring frames to re-synthesize
the interpolated frames. This is what additive synthesizers do.
Morph – Spectral Same as Morph - Spectral, but the phase content of the fundamental is
(Zero Fund. Phase) zeroed for all source frames. This way the lowest frequency does not shift/
rotate between frames.
Morph – Spectral Same as Morph - Spectral, but all phase content is discarded. This might alter
(Zero All Phases) the sound of the source content drastically, and therefore can sometimes be
undesirable.
However this option also creates the smoothest transitions between frames
since no frequencies need to shift phase.
Remove Morph Tables Revert back to how things were before interpolation was applied.
Note that clicking “Undo” might be the better choice for reversing the morph
since, in the case of the spectral (zero-phase) modes, the two zero-phase
choices destructively alter the source tables.
Saving Wavetables
If you have modified one or more frames in the Wavetable Editor, the wavetable name changes to a
tinted background indicating that the wavetable has been modified but not yet saved to storage.
Wavetable Modified
Right-click Custom and choose Save Table in the menu that appears.
Choose the file name and location in the dialog that appears.
Note that you do not need to save your wavetables explicitly. Serum
always saves changes you make to wavetable data inside a preset (your
song) unless it is a factory wavetable. This uses some hard disk space
— how much depends on how many frames you use in the wavetable,
from 8k to 4 Megabytes
The benefit, however, is that you can exchange presets with others, or
open your song in the future, without having to worry about table file
management. The only reason to explicitly save a wavetable is to have it
appear in the Wavetable menu.
Saving a Wavetable
It’s important to realize that wavetables do not behave like samples in a sampler. When importing
audio to create a wavetable, you shouldn’t expect to create ultra-realistic reproductions of acoustic
instruments. Use Sample mode (in any of the oscillators) to work directly with source samples.
Samplers typically play audio as a single stream of data. In contrast, when importing a sound into Serum,
it attempts to slice the sound into individual single-cycles. Because of the nature of audio, it’s generally
best to select source sounds that are monophonic (that is, sounds that contain a single pitch). This
means that, in most cases, a single note is better than a chord.
Because these single cycles become the basis of the oscillator, pitch information is effectively removed.
In other words, if you load a sound that has a pitch bend, the pitch bend will no longer exist. However,
wavetables have their own set of advantages, which one could describe as a “solid” or “fixed” sound that
lends itself well to unison and wave manipulation (sync, FM, and more) without sounding flimsy.
Serum therefore does a fantastic job of importing the waveform of other sounds. Fixed-pitch, one-
shot (monophonic) sounds, such as a one-shot sample of a synthesizer, are among the best choices for
importing into Serum. Speech and other complex sounds can however yield some pretty interesting
results if you’re open to some experimentation.
Note: Dragging files directly from the host arrangement window or “region bins” is not possible in most
hosts, but using the host’s standard file-browser should work. In addition, many hosts are able to show
the (parent) sound file in the host file browser, and you should be able to drag from there.
When importing stereo files as a wavetable, Serum uses the left channel in OSC A and
OSC C, and uses the right channel in OSC B.
As you drag the waveform, you’ll see several options appear. The location where you release the
mouse determines the import method. Serum then analyzes the sound and creates a new wavetable in
memory. The analysis specifics depend on the import method you choose.
Serum scans the audio file and builds a pitch map. Serum then attempts to locate zero-crossings
within the pitch map. While this works well with simple sounds, complex sounds typically don’t
adhere to sensible zero crossings, so you’ll end up with at least some glitches at best.
Use this mode when you have a sound with a non-fixed fundamental (pitch bend or vibrato) and
the sound is pretty simple, for example a sawtooth wave with little filter sweep/resonance.
Similar to the previous option, Serum builds a pitch map and imports a varying-sized segment of
audio for each frame (subtable) based on the analyzed pitch.
Unlike the previous option, pitch follow import does not attempt to locate zero crossings. This
option is therefore better suited for complex sounds, such as a source sample that might have a
touch of chorus/unison, resonance, or background noise/notes.
• FREQUENCY ESTIMATION
Serum analyzes the incoming waveform to determine its dominant frequencies and harmonic
content, and uses this information to convert the audio into a wavetable. This involves analyzing
the audio signal to identify the fundamental frequency, and then determining the harmonics
(integer multiples of the fundamental frequency).
By matching the detected fundamental frequency to a musical pitch (for example, C4 or A3), Serum
can align the wavetable content to the correct pitch. Serum then uses the estimated frequencies
to guide the conversion of the audio signal into a series of wavetable frames. This ensures that
the harmonic structure and frequency content of the original audio are preserved within the
wavetable, maintaining the character of the original sound.
Frequency estimation is particularly beneficial for audio samples with well-defined pitches and
harmonic structures. These types of samples allow Serum to accurately extract fundamental
frequencies and harmonics, making it easier to convert the sounds into expressive and playable
wavetables. Examples include monophonic synth leads, bass sounds, vocals with sustained notes,
plucked and struck instruments, and FM (frequency modulation) generated sounds.
This option is typically the best choice when a sound has a fixed frequency, such as a one-shot
from a synthesizer (in other words you hear it as a perfect or near-perfect constant pitch, with
essentially no pitch bend or vibrato).
In this mode, Serum analyzes the entire file for an average pitch, and then uses this number of
samples as the import length. Because some sounds contain half-cycles, silence, multiple notes,
and other artifacts, Serum might not correctly guess the desired pitch.
Fortunately, Serum displays the number of samples it is using per frame in the Wavetable Editor
formula area and switches over to this “fixed” value found during analysis (unless changed or
cleared from the Formula field).
• FFT 256/512/1024/2048
FFT, or Fast Fourier Transform, is a method that converts a signal from the time domain to the
frequency domain, revealing its frequency components. It efficiently breaks down a complex
waveform into individual sinusoidal frequencies, enabling analysis of amplitude and phase at
different frequencies.
Unlike the other import modes, which import and divide (and possibly stretch) chunks of the
original waveform, the FFT modes are a spectral import. This means that these import modes
divide the source audio into small snippets of time, and analyze the spectral content.
One way of thinking of this is a “blurred averaging of the frequency content”. This can be very
useful for sounds such as drum loops, speech, and other material where you want the flavor of the
sound for abstract purposes.
The numbers 256, 512, 1024, and 2048 represent the number of samples used to perform the
FFT analysis. FFT 256, for example, analyzes 256 samples at a time.
Larger FFT sizes, such as 1024 or 2048, provide higher frequency resolution because more
samples are considered during the transform. This means you get a more detailed breakdown of
frequency components, making it better for detecting specific tones and harmonics.
Smaller FFT sizes, including 256 and 512, provide lower frequency resolution, which results in a
broader view of the frequency spectrum. This can be less detailed but faster in terms of processing.
You can choose to import the audio as a regular sample, automatically switching to the Sample,
Granular, or Spectral oscillator modes. In these cases, the audio is imported directly without
creating a wavetable.
Advanced Imports
Due to the complex nature of audio signals, using a pitch average might not always be perfect.
Sometimes you might want to specify the exact number of samples for each cycle (that you can measure
yourself in a sample editor).
The Wavetable Editor includes a formula parser (described in the next chapter) featuring a text box with
the placeholder text “(enter formula)”. Although the primary function of this field is to create waveforms
from functions, you can also type the following:
• A one to four-digit number (such as 1024) to instruct Serum to split the sound file into segments
with this number of samples.
Serum converts the sound file into the appropriate nearest number of samples (rounding to the
nearest whole number of samples).
This MIDI note/sample conversion assumes a 44,100 Hz source sound file. If your source sound
file is not 44,100 Hz, the number of samples is still held true, but you should disregard the
displayed frequency/note information in the formula text area.
Note: Since musical note frequencies are not often perfectly divisible by a sample rate, many pitches will
have some visible “drift” on import.
When creating a wavetable from imported audio, you might hear undesirable buzz from
subharmonics or other factors caused by the beginning and end of the wave cycle not lining up
perfectly.
Therefore, after importing using one of the above methods, try using a fade command from the
Process menu. For example, X-Fade Edges (16 Samples) or X-Fade Edges (Grid Size) should
give you a less buzzy-sounding result.
For example, if your audio file is sound.wav, Serum looks for the sound.txt file. If Serum can’t
find the matching file, Serum then looks for the FolderInfo.txt file in the same folder. The
FolderInfo.txt file contains information that applies to all sounds in the folder; this saves you from
having to duplicate the same text file for every WAV file in the folder.
The first line specifies the number of samples-per-cycle of the audio file (in square brackets). Valid
values are numbers between 32 and 9999. The second line specifies whether to create intermediate
interpolated tables. The default is yes; use [no interp] to specify the opposite.
Internally, Serum uses 2048 samples for each single-cycle. This makes 2048 the ideal
number of samples to use as a wave cycle if you plan on generating sounds from
synthesizers, vocoders, and so on, to import into Serum (because no resampling of the
source audio is needed).
This 2048 samples-per-cycle works out to 46.875 Hz at 96000 KHz sampling rate
(96000/2048 = 46.875), which is F#0 +24 cents. Since the octave might differ in various
hosts and synthesizers, render and measure in a program that allows you to select in
samples, such as the freeware sound editor Audacity).
Type 2048 in the Formula field prior to dragging to import. Serum will then not have to
alter the sample data at all.
This replaces the waveform in the specific frame; the rest of the wavetable remains untouched.
To import multiple single-cycle waveforms at once, drag them together to the oscillator waveform view.
Whenever possible, use low-frequency notes. Also, use 2048 samples-per-cycle, which
precludes the need for Serum to interpolate the sound.
Alternatively, use low C at 44,100 Hz (33 Hz, 1349 samples per cycle). This is slightly easier
to set up since no fine tuning of pitch is needed.
The following procedure shows an example of how to export a 2048 samples-per-cycle wave from
another synthesizer and import it into Serum.
Open the Serum Wavetable Editor and type 2048 into the Formula text box. Serum displays the
following in the field:
split at: 2048 samples (22 Hz, note: F1 and -22 cents)
This suggests that you play a low F with -22 cents to create the proper length of 2048 samples per
cycle.
Do the following:
a. Create the F1 note in your host sequencer piano roll on the synthesizer track.
Note that octave naming might be different since, unfortunately, there is no standard. You should
probably use a spectrum analyzer to verify ~22 Hz is the fundamental, not ~11 Hz or ~44 Hz.
As for the duration of the note, since Serum can handle 256 wave cycles, you can render this
22 Hz file for up to 11.6 seconds, which is about 8 bars at 165 BPM. If, on the other hand, the
sound has very little modulation/change, there is no need for such a long note. You can probably
determine by ear when the source sound has stopped changing; there is no need for a longer note.
How you do this varies depending on the synthesizer. Typically, this is set per-oscillator, but some
synths have a global tuning.
If you can’t do this in the source synth, consider adding the appropriate amount of “pitch bend”
down as a MIDI message (22% down if pitch bend range is 1 on the source synth). A=440 to
approximately A=434.5 Hz is another potential option.
c. Render (export) this note from the synth as audio. Alternatively, you can freeze and flatten to
produce the synth note as an audio file.
If the value 2048 is no longer visible in the Formula text field (for example, if you closed the Serum
window after starting this procedure), type 2048 into the field.
When you release the mouse, the audio is imported. The source audio file (render) is no longer
needed at this point unless you wish to re-import it into Serum with different settings.
• You will likely have blank or unnecessary tables at the end of the wavetable.
Using the thumbnails, find the first table you don’t want to include. Click the button
and choose Remove: Selected->End in the menu that appears.
• You might want to apply fades to the tables (use one of the Fade or X Fade options using the
Wavetable Editor Process menus) to eliminate any buzzing artifacts from frame edges.
• Listen to the entire wavetable set by playing a note and dragging the WT POS number box in the
Wavetable Editor (or, similarly using the WT POS knob on the main panel for the oscillator).
Alternatively, for automatic playback, you can use a mod source (for example, an envelope with
long attack or an LFO with an upward saw shape) to modulate the wavetable position.
To automate a sweep
through the wavetable,
draw an LFO ramp and
drag the LFO tab to the
WT POS knob.
LFO Ramp
This means that if you move the audio sample to a different location on your drive, Serum will not be
able to locate the file automatically. When you try to load a preset that has missing audio files, Serum
displays a dialog similar to the following:
• Copy the missing audio file to a folder within the current search path and click the
button.
• Click the button to include the folder containing the audio file in the search path, then click
.
To simplify file management, Serum also allows you to embed the wavetable generated from the
imported audio into the preset when saving.
Specifically, the Formula parser field allows you to enter math functions to plot tables in a generative
fashion. This is a fairly advanced feature; don’t worry if you don’t follow all the details. You can still
explore this capability through experimentation.
However if you want to generate waveforms from thin air and enjoy this kind of approach, you’ll be glad
that you paid attention in math class!
The good news is that Serum includes many formula presets, and it’s easy to paste formulas created by
other users.
Basic Functions
The following table lists the basic functions supported by the parser.
• sel refers to the current (existing) waveform value of the selected table
This does not change for each table when using a multi-table formula. In other words, the formula
contains y or z, and refers only to the selected table when formula processing begins.
Symbol Description
pi 3.141592658979323846264338
e 2.718281828182818281828
w The current time-value getting plotted, from 0.0 to 1.0. This is the same as
(x+1)/2.
x The current time-value being plotted, from -1.0 to 1.0.
y The current table number, from 0.0 to 1.0.
z The current table number, from -1.0 to 1.0. This is equivalent to (y\*2)-1.
q When a q is present in the formula, the function plots to the FFT bins
instead of plotting to the waveform display. Note that q iterates from 1 to
512 for the respective harmonics/bins.
in The current (old) visible waveform value of the plotting table. This changes
to each old table, if using y or z, which plots all tables.
sel Similar to in, but only the currently-selected wavetable (does not change to
each table; uses the selected table when formula processing begins).
rand A random number from -1.0 to 1.0 that stays the same for all tables
(precalculated for every time position).
• y and z refer to the current table number. Therefore, when y or z is in the expression, all tables for
the current oscillator are regenerated.
• q renders the function to FFT instead of to the waveform display. There is no reason to include x
or y in a formula containing q.
As soon as you select a formula, it appears in the Formula field automatically and the waveform is
calculated.
Singles These are single-cycle formulas that do not contain y or z variables. When
you select an option, only the currently-selected table is replaced/generated
with the formula.
Multis These are multi-cycle formulas that contain y or z variables. When you select
an option, the entire wavetable is replaced.
User Singles These are user-defined single-cycle formulas.
User Multis These are user-defined multi-cycle formulas.
1. Type the formula in the Formula field and press the Enter key.
2. Click the formula presets menu and choose Save Formula in the menu that appears. A dialog
appears allowing you to save the formula.
The formula is saved in the corresponding formula sub-menu (either User Singles or User Multis,
as appropriate). You can now access your formula using the formula presets menu.
• FormulaUserMultis.txt
• FormulaUserSingles.txt
You can find these files in the Serum 2 Presets/System/ folder. The text files have the following
format:
[formula1][formula1-name]
[formula2][formula2-name]
If you need to remove a formula preset that you saved earlier, do the following:
1. Open the appropriate user formula preset file using a text editor.
This means you can do signal generation or processing of a wavetable in the frequency domain. In this
case, the q value represents the bin number in the FFT area at the top of the Wavetable Editor, from 1
to 512.
q<17
The formula creates a wavetable frame with only the first 16 harmonics.
Formula as q<17
q<z*256
This formula creates an entire table set of 256 frames, with one additional harmonic in each
consecutive frame.
(1/q)^0.25
This creates a sawtooth wave with higher harmonics than a normal saw (-3 dB/oct). Replace the 0.25
value with a different number to alter the spectral decay (for instance, 0.5 produces a normal saw, 0.75
creates a duller saw).
(1/q)^z
This causes Serum to create 256 frames in the wavetable with the harmonic scale.
((q%2)==1)?in:0
This formula removes every second harmonic (even harmonics) from the current wavetable frame,
leaving odd harmonics only (like a square wave).
Formula as ((q%2)==1)?in:0
Using this on a (default) sawtooth wave results in a square wave. This is a popular formula, particularly in
Dubstep bass sound design, to make tables sound more “hollow” (taking up less of the spectrum).
There are two types of global settings: those saved with your preset, song, or patch, and those saved as
part of Serum, accessible to all presets.
The page is divided into the following panes to help you quickly find the appropriate settings:
• Preferences — Specify global preferences including user interface and MPE settings
• Voice Control — Define the behavior of each voice across the available oscillators
The settings page also displays the current Serum version and build date.
Preferences
Use the PREFERENCES pane
to configure your global
preferences in Serum.
User Interface Help Tooltips Display tooltips after hovering the mouse
pointer over a control for a moment.
Param value tooltips Display numeric values (as a small pop-up)
when modifying a control.
Double-click params Specify whether double-clicking a control
resets the control to the default (init) value
or whether double-clicking displays a pop-up
text box allowing you to type a specific value.
User Interface Mouse wheel param control By default, moving the mouse wheel adjusts
(cont.) the knob that the cursor is currently hovering
over. If you don’t want this capability, enable
this setting to deactivate the feature.
Keyboard shortcuts Set whether Serum should respond to input
from the computer keyboard. Disable this to
prevent Serum from stealing keyboard focus
from your DAW.
Default waveform view The default waveform view in the OSC
panels, either 2D (default) or 3D.
MPE MPE enabled by default Specify whether MPE mode is enabled when
a new instance of Serum is added to your
project.
MPE Pitch Bend (also) By default, Serum maps MIDI pitch bend
maps to Expr X messages to per-note pitch bend and CC10
(pan) messages to note expression X.
General (cont.) Mod Wheel -> WT Pos When the Wavetable Editor is active, the
(when WT Editor is open) mod wheel scans the indices from 1-256.
Silence note + FX tails when Mute effects and any sustaining notes when
host transport stops the host DAW is stopped.
Load MIDI Map from Normally, Serum loads a default MIDI CC
Presets map if the Serum 2 Presets/System/
MIDI CC Maps/Default.SerumMIDIMap is
found, in the following cases:
Loading a Preset
You can optionally load a factory
supplied or user-defined preset.
Setting Randomization
You can set the randomization for
the PAN, DETUNE, CUTOFF, and
ENVS (envelopes).
Setting Randomization
Field Description
You can use the arrow keys to fine tune the setting. You can also double-click the field and type a value.
To set the LFO scaling, click the field and drag up or down. You can use the arrow keys to fine tune the
setting. You can also double-click the field and type a value. Choose to set by percentage or rate.
Setting Description
S1 Compatibility Serum 2 features a completely rebuilt sound engine. However, when you
Mode load a Serum 1 preset, this option is automatically enabled to preserve
maximum sonic similarity with Serum 1.
Disable this option if you prefer that Serum 1 presets instead take advantage
of the DSP updates available in the Serum 2 sound engine.
Setting Description
To set the concert pitch for Serum, click the left field and drag up or down. You can use the arrow keys
to fine tune the setting. You can also double-click the field and type a value.
Using the free MTS-ESP MINI plugin, you can load .scl, .kbm or .tun files and automatically retune
all connected MTS-ESP clients (including Serum). This allows you to retune any number of supported
virtual instruments from a central location without requiring you to tune each instrument separately.
MTS-ESP support is enabled in Serum by default. You can disable this feature by unchecking the Enable
MTS-ESP menu option.
Enable the MTS-ESP Note-On Only option to have the tuning set on Note-On MIDI events. This
ensures that the tuning of a note will not change during its duration, even if the global MTS-ESP tuning
updates.
• Loading a .tun file always takes precedence over MTS-ESP. This allows you to tune any Serum
instance differently from the global MTS-ESP tuning.
• If the oscillator pitch mode is set to Steps (set by right-clicking an oscillator OCT or SEM control),
you can pitch oscillators up or down in “periods” and “steps” as defined by the active MTS-ESP
tuning, rather than in octaves or semitones.
In addition to the free MTS-ESP MINI plugin, you can choose from among the following additional
MTS-ESP plugin options:
• Wilsonic MTS-ESP
• Surge XT
• Entonal Studio
• Infinitone
This means that when locked, Serum ignores the TUN file and the concert pitch setting in any new
preset that you load.
You can access the main menu near the top right of the Serum window.
Serum 2 Presets/Presets/User/
default.SerumPreset
Note that if you are running the Serum FX version,
this saves to the defaultFX.SerumPreset
file. This allows you to configure different default
configurations for Serum and Serum FX.
When you load Serum FX, it will look for the
defaultFX.SerumPreset file. If the file isn’t
found, Serum FX instead looks for the
default.SerumPreset file.
Import Import Preset Pack Import a Serum preset pack. A system dialog appears
allowing you to locate the preset pack.
Rendering Render OSC Warp Use the current wavetable frame of the selected
oscillator and create 256 frames (subtables) spanning
0 to 100% of the WARP knob.
Resample to Play a note of the preset for one bar and capture
(render and import) the result as a wavetable in the
selected oscillator (or OSC A and OSC B).
Folders Open Serum 2 Presets Display the Serum 2 Presets folder using the Finder
Folder (macOS) or Explorer (Windows).
Rescan Folders on Rescan the Serum 2 Presets folder.
Disk
Do this when you make changes to the folders
outside of Serum (using the Finder or Explorer).
MIDI/Tuning Load MIDI Map Load a saved MIDI CC map from storage.
Save MIDI Map Save a MIDI CC map to storage. By default, MIDI
maps are stored in the Serum 2 Presets > System>
MIDI CC Maps folder.
Saving a MIDI map as default.SerumMIDIMap in the
default folder instructs Serum to load that MIDI map
every time you add an instance of Serum or you load
a preset.
Load Tuning (.tun) Load a tuning file for the current instance of Serum. A
dialog appears allowing you to locate the appropriate
tuning (.tun) file.
See “Using a Tuning File” for more information.
MPE MPE Enabled Enable support for MIDI Polyphonic Expression
(MPE).
When enabled, Serum responds to MPE messages,
allowing for more expressive and nuanced musical
performances using compatible MPE controllers.
Important: In the VST3 version of Serum (not the AU
or AAX versions), when MPE is disabled, Serum will
respond to VST3 Note Expression in cases when the
host DAW supports this feature (this includes hosts
such as Bitwig, Cubase, and Nuendo).
This means that main menu options to map note
expressions to macros remain available and applicable
even when MPE is disabled. However, the MPE Bend
Range option is not available since VST3 pitch note
expression has a fixed range of +/-120 semitones.
MPE: XYZ -> Macro Map the X, Y, and Z axes of an MPE-compatible
1,2,3 controller to Serum macros 1, 2, and 3 respectively.
In this context, the x, y, and z axes represent the three
dimensions of touch-sensitive control associated with
MPE.
Specifically:
MPE (cont.) MPE: YZ -> Macro 1,2 Map the X and Y axes of an MPE-compatible
controller to Serum macros 1 and 2 respectively.
MPE: Y -> Mod Wheel Map the Y-axis movement of an MPE controller to the
Modulation Wheel (Mod Wheel) control.
MPE Bend Range: 48 The pitch bend range for MPE controllers, specified as
the number of semitones above or below the original
pitch.
Choose this menu option to display a dialog allowing
you change the current value. You can set any value
from 1 to 96 semitones.
A wider range allows for more expressive pitch
variations and glides across the tonal spectrum.
Click the button to access the presets browser. The browser appears.
This appendix explains how to use the presets browser to quickly locate the sounds you need and
organize your favorites in a way that best suits your workflow.
Presets Folders
Select a folder in the hierarchy to have the folder contents (presets) appear in the Presets pane.
Loading a Preset
Click an entry in the list to load the corresponding preset.
Most presets load immediately; presets with larger embedded samples, such as multisampled
instruments, display a small green progress bar (directly beneath the preset name) when loading.
Previewing Presets
The browser makes it easy to preview presets. Click the corresponding button to hear the preset
play.
Serum also allows you to auto-preview clips. See “Performing Standard Preset Operations” on page
334 for more information.
Preset Playing
After selecting a preset, press the right arrow key to start the preview and press the left arrow
key to stop the preview. Use the up and down arrow keys to quickly change presets.
While previewing a preset, you can modify any macros that are defined as part of the preset and hear
the effects immediately.
Preset Macros
Searching Presets
You can use the presets browser to help locate the right type of sound by name, category, tag, rating,
and more.
Searching by Name
Perhaps the easiest way to find presets is to search by name. Preset names often contain important
sonic attributes as part of their name. For example, if you’re searching for variations of piano sounds, it’s
often the case that the preset name will include “piano” as part of its name.
The search results appear. Click the x button to clear the search results.
Press Cmd-F (macOS)/Ctrl-F (Windows) to position the cursor in the search field. Note that
this only works when you are already in the presets browser.
Searching by Ratings
You can search for presets based on the ratings you’ve assigned.
Note: Ratings are user-assigned; presets, by default, have no assigned rating. Ratings allow you to
quickly score presets that you find interesting, allowing you to easily locate them later.
Click the Ratings menu and choose a score from the list
that appears.
Click a preset in the list. The preset highlights and the metadata appears in the pane on the right.
Preset Metadata
Modify the information, as appropriate. In most fields, click in the field and edit the contents. Two fields
are slightly different: CATEGORY and TAGS. You can read about those below.
Note that you cannot change the DATE field; the preset creation date is managed by the computer
operating system and represents when the preset file was created (or last modified).
The new category now appears in the list for you to use later, if
needed.
Preset Category
Managing Tags
You can manage the tags associated with a preset.
Preset Tags
Rating Presets
Preset Tags
You can rate presets using a one to five-star system and easily search for presets based on their ratings,
allowing you to quickly find presets that capture your interest.
Rating a Preset
Click the corresponding star for the preset you want to rate. To change a rating, click the new star
rating.
To remove a rating, click the first star to select it and then click the same star again to remove the rating.
Presets Menu
Presets Show Preset in Finder Display the selected preset in the Finder (macOS) or
(macOS) Explorer (Windows).
Show Preset in Folder
(Windows)
Load Random Preset Load a random preset. To load another random
preset, type 7
Hybridize Load a hybrid preset consisting of four randomly-
selected presets. To load another hybrid preset, type
8.
The generated preset name reflects the hybrid mix.
If you accidentally delete one or more presets, Serum moves the preset files to the Trash
(macOS) or Recycle Bin (Windows).
Using macOS, open the Trash, find the preset file, right-click the file, and choose Put Back from
the menu.
Using Windows, open the Recycle Bin, find the preset file, right-click the file, and choose
Restore from the menu.
Return to Serum, click the button in the presets browser, and choose Rescan Database
from the menu. The presets reappear.
You can also perform the most common preset operations by right-clicking a preset and choosing an
option from the menu that appears.
Preset Menu
Operation Description
Show in Finder/Folder Display the selected preset in the Finder (macOS) or Explorer (Windows).
Delete Delete the selected preset.
Important: Serum does not ask for confirmation before deleting, and there
is no way to undo the deletion from within Serum. Don’t despair. See the
earlier tip to recover any accidentally deleted presets.
Rename/Move Rename or move the selected preset.
A macOS or Windows system dialog appears. Use the dialog to rename or
move the preset, as appropriate.
After returning to Serum, click the button in the presets browser, and
choose Rescan Database from the menu.
Field Description
PACK NAME The name of the pack. This is used as the default file name for the preset
pack.
AUTHOR The author name.
Make sure that the button is enabled to prevent users from modifying
the AUTHOR field.
URL The URL of your website.
DESCRIPTION A description of the preset pack.
PACK ARTWORK The artwork for the preset pack. The ideal image size is 436 x 216 pixels.
Click the button. A dialog appears allowing you to locate the image file.
Init disabled Reset disabled oscillators, removing associated parameteres and samples.
oscillators
Click the button. A dialog appears allowing you to specify the file name and folder.
The dialog displays details about how the pack was created.
Click the button to show the location of the artist pack. The name of the file is:
<pack-name>.SerumPack
• (macOS) ~/Library/Preferences/
This file stores the preferences accessible on the Preferences page, the last known path to the Serum
Presets folder, and (optionally) power user options, among other options.
Serum automatically recreates this file if it is missing, so you can reset your preferences to the
factory defaults simply by deleting this file and restarting Serum.
You probably shouldn’t change many of the settings. However, there are a few power user settings that
you might want to explore.
Important: When editing the JSON file, use a text editor application such as TextEdit (macOS) or
Notepad (Windows). Do not use an application that saves the file in any format other than text (the
standard format for JSON files).
Very Important: Each line in the configuration file ends with a comma (,). Ensure that you keep the
comma at the end of each line when you edit the file.
“Default Author”: “ “,
Type the name you want to use between the quotation marks. For example:
Any presets made from scratch will now have the name you specified listed as the artist.
“Enable CCForRockers”: 0,
“Enable CCForRockers”: 1,
The first line maps the preset forward arrow (>) to a MIDI CC number; similarly, the second line maps
the preset backward arrow (<) to a MIDI CC number.
Update the -1 values on each line to the appropriate MIDI CC number for your controller (button).
For example, if you would like to assign MIDI CC 21 to the preset forward arrow and MIDI CC 22 to the
preset backward arrow, change the lines to the following:
For the settings discussed in this section, the value -1 means unassigned.
To remove the MIDI CC assignments that you configured in this section, edit the file and reset
the corresponding values to -1.
A MIDI controller button mapped to a CC typically sends a value of 127 when pressed and
a value of 0 when released. In Serum, a value of 64 or above triggers the action (changes
the preset). Similarly, Serum needs to receive a value below 64 before the action can be
triggered again.
“Enable CCForRockers”: 0,
The value on this line might already be 1 if you followed the procedure in the previous section. Change
the line to the following:
“Enable CCForRockers”: 2,
Again, the values might be different than -1. As before, the first line maps the preset forward arrow (>)
to a MIDI note number; similarly, the second line maps the preset backward arrow (<) to a MIDI note
number.
Update the values on each line to the appropriate MIDI note number.
“Enable CCForRockers”: 0,
“Enable CCForRockers”: 1,
The first line maps the OSC A preset forward arrow (>) to a MIDI CC value; similarly, the second line
maps the OSC A preset backward arrow (<) to a MIDI CC value.
Update the -1 values on each line to the appropriate MIDI CC value for your controller (knob).
For example, if you would like to assign MIDI CC 23 to the OSC A preset forward arrow and MIDI CC
24 to the OSC A preset backward arrow, change the lines to the following:
Remember to include the comma (,) at the end of each line. This is important. Also, recall that you can
reset these settings by changing the values back to -1.
Note that OSC B, OSC C, and OSC N (the noise oscillator) have similar settings in the configuration file.
Use the same procedure to assign those to appropriate MIDI CC values for your controllers (knobs).
The Serum Presets folder contains all the files that Serum reads and writes, except the preferences file.
Serum installs this folder to the following locations by default:
• (Windows) /Documents/Xfer/
The following describes some of the more important folders in the file structure:
Folder Description
Folder Description
LFO Shapes Contains LFO shapes (.XferShape files) that are in the same file format
as the shape files used in the LFOTool plug-in. LFO shapes appear in the
following locations:
• The waveshaper (in the FX section, when the X-Shaper effect type is
selected and the waveshaper graph is displayed)
• The Remap editor (when WARP mode is set to Remap and the graph is
displayed)
Presets Contains subfolders holding the presets that you see in the presets browser
and menu. These include the factory presets together with any presets you
save.
Samples Contains factory-supplied (tonal and non-tonal) and user-saved samples.
System Contains the four formula files:
• FormulaFactoryMultis
• FormulaFactorySingles
• FormulaUserMultis
• FormulaUserSingles
You can edit any of these text files, but the intent and recommendation is
to edit User files and leave Factory files untouched. See “Using the Formula
Parser” on page 304 for more information about formulas.
The System folder also contains a MIDIccMaps folder, which is where any
MIDI CC maps you create are stored. Finally, the User.dat file holds your
registration information.
Tables Contains subfolders holding the wavetables that you see in Serum. These
include the factory presets along with any wavetables you save. Wavetables
are special Serum-saved WAV files.
You can create your own subfolders in the Tables folder, but Serum does not
scan deeper (no sub-subfolders are scanned).
This typically is a sound with a lot of frames, but ones that fluidly work together (think of a pluck
string that decays across many cycles). In general, there might be a lot of individual cycles, but they
all go together well and relate nicely with each other.
Another example would be a synthesizer sample. You want many cycles so it doesn’t feel too
“steppy,” however all the frames should feel as part of a collection (a similarity/sweep across them).
Many of the factory presets in popular wavetable synths contain only four or five cycles. This
provides some variety within the table, but not so many waveforms that it feels disconnected or
crazy.
Table Ordering
It typically makes sense to have the frames progress from dull to bright, or vice-versa. You may have a
situation where you want the spectrum to peak somewhere in the middle, but probably not.
You can drag the thumbnails at the bottom of the Waveform Editor to rearrange them. This way, when
you move the WT POS knob, it feels as if you’re traveling in a straight line, instead of in a zig-zag
fashion (spectrally speaking).
Interpolation
If you have four frames and automate the WT POS knob, you’ll hear four discrete tones immediately
jumping from one to the next. This is typically undesirable.
In the Wavetable Editor, you can select MORPH > Morph - Crossfade. This is what many wavetable
synths do automatically in every preset wavetable. However, there are times when you might prefer
discrete (non-interpolated) waveforms.
• Drawing
The grid size in lower-right determines snapping. Try different sizes such as 6 or 12 to bring out +7
and +19 harmonics.
Right-click in the bins for a pop-up menu with more options (including random, among others).
Use the curved arrow near the top left and add a new table (using the > button in the lower left).
Then make some adjustments and repeat.
About four frames seem to be a popular number in many other software synthesizers.
Managing Unison
Unison is a powerful tool for enhancing the depth and richness of a sound by layering multiple voices
slightly detuned or panned, creating a fuller, more powerful tonal presence. However, overuse of unison
can potentially lead to quality and performance issues.
Using more than three to seven unisons per oscillator is often unnecessary and can potentially
negatively impact efficiency. This is because higher unison counts not only significantly increase
CPU usage, but can also introduce phasing issues, potentially degrading your sound quality.
Instead of stacking unisons to create a thick, chorused sound, use a dedicated FX bus with a
chorus effect. This approach offers two principal advantages. First, by apply the effect once instead
of processing it for every voice, this approach is considerably more CPU friendly.
Second, this provides greater flexibility by allowing easier tweaking and layering of effects without
duplicating processing effort.
The copy module shortcuts also apply to oscillators on the MIXER page.
Sample/Granular/Spectral
Audio
FX
LFOs
Modulation
Wavetable Editor