Antares Harmony EFX user manual

							5
Chapter 2: Introducing Harmony EFX
What the heck is this 
thing?
Harmony EFX is an extremely 
affordable and easy-to-use version of our 
Harmony Engine Evo real-time harmony 
generating plug-in. Based on exactly the 
same Evo™ Voice Processing Technology 
that powers its big brother, Harmony EFX 
puts professional-quality vocal harmony 
arrangements within reach of any songwriter, 
producer, musician or engineer.
What’s it do?
Traditionally, harmony tools offer two basic 
operating modes:
o n one hand are fully automatic modes based 
on either parallel or “smart” scale-based 
intervals. These are relatively easy to set 
up, but offer very little in the way of creative 
control.
o n the other hand (assuming organisms with 
two hands) are MIDI modes that provide 
control of each individual note of each individual 
harmony voice. While these techniques offer 
absolute control, they also demand that you be 
skilled in the intricacies of vocal arranging (as 
well as having the time and patience to enter 
every note into a MIDI track, hardly a trivial 
undertaking).
With Harmony EFX, we set ourselves the task 
of creating a tool that would let anyone who 
could hear the harmonies they wanted in their 
mind quickly and easily create those harmonies 
in a song or other project. While Harmony EFX 
offers all of the traditional harmony generation 
methods, it adds a variety of new operating 
paradigms that allow you to approach the 
harmony generation process from a purely 
musical point of view.
What’s it got?
With four independent harmony voices 
utilizing our advanced Evo™ Voice Processing 
Technology, a variety of powerful harmony 
generating modes, humanization features for 
natural sounding performances, and a flexible 
real-time harmony preset system, Harmony 
EFX provides incredibly easy-to-use tools to 
quickly and easily produce virtually any vocal 
arrangement you can imagine. 
Key Harmony EFX features include:
•  Four high- quality, formant- corrected harmony 
voices featuring Evo™ Voice Processing 
Technology with independent vocal character 
and pan settings
• Antares Throat Modeling technology that lets 
you process each harmony voice through a 
physical model of the human vocal tract
• A variety of innovative Harmony Control 
modes, from fully automatic to individual 
control of every note:
   Fixed or Scale Intervals: Simply set the 
key and harmony voice intervals and let 
Harmony EFX do the rest
   Scale Degree or Chord Name: Define 
your harmony chord-by-chord, complete 
with inversions and variable vocal ranges
   Chord by MIDI: Define your harmony 
in real time via a MIDI controller or pre-
recorded MIDI track
   MIDI Omni:  Directly “play” the four 
harmony voices as if they were voices of 
a synth or sampler.
   MIDI Channel:  Use four separate MIDI 
channels for absolute control over each 
note of your vocal arrangement
•  Spread and Register controls that allow you 
to quickly and intuitively set the pitch range 
and harmony style of your vocal arrangement 
						

							6
• Humanize functions that provide selectable 
amounts of variation to each harmony voice 
for realistic, natural sounding results
• A Harmony Preset matrix that lets you create 
up to 12 complete harmony settings and 
instantaneously recall them, in real time or 
via automation
Whether you’re an experienced vocal arranger, 
a songwriter looking for that perfect backup 
vocal, or a composer experimenting with 
unique vocal effects, Harmony EFX gives you 
entirely new ways to create the harmony parts 
you hear in your head. In fact, experimenting 
with different harmonies is so easy (and, dare 
we suggest, fun), you may find yourself using 
Harmony EFX to explore harmonic alternatives 
you may have never otherwise considered.
Harmony EFX is, quite simply, the quickest, 
easiest tool for flexible, realistic harmony 
creation. 
  
						

							7
Chapter 3: Harmony EFX Controls
This chapter is a reference for all of the controls used in the Harmony EFX interface. 
Controls
Continuous controls in Harmony EFX are 
represented by virtual knobs or faders. 
Knobs can be controlled by either vertical or 
horizontal movement: 
VERTICAL: Position the cursor over a knob, 
press and hold the left mouse button (or 
the only mouse button, if you’re using a 
one-button mouse) and move the cursor 
up to turn the knob clockwise or down 
to turn the knob counterclockwise. The 
current value of the knob’s parameter 
appears in its associated numeric display.
H o RIZ oNTAL: Just like Vertical, but move 
the mouse right for clockwise rotation 
and left for counterclockwise rotation.
For faders, simply click on the “thumb” of the 
fader and move it to the desired setting. o
ption (Mac)/Control (PC)-clicking a control 
returns it to its default value. 
N OTE : The specific modifier 
key may vary from host to 
host (Steinberg hosts on the 
Mac, for example, use the Command 
(Apple) key instead of the Option key). 
Input Audio
For accurate harmony generation and best 
modeling performance, Harmony EFX needs 
to be able to detect the pitch of the original 
performance. To do that, Harmony EFX 
requires a clean, pitched, monophonic signal. 
(In this context, “clean” refers both to lack of 
noise and lack of processing with effects like 
chorus, reverb, etc.) 
						

							8
NOTE : Keep in mind that the input 
must not only be monophonic, but 
actually a solo voice or instrument 
(as opposed to a unison section). For example, 
a group of singers singing the same note will 
typically not be reliably tracked. 
Although even true solo voices can sometimes 
be challenging (for example, exceptionally 
breathy voices, or tracks recorded in 
unavoidably noisy environments), Harmony 
EFX’s Tracking control, described later in this 
chapter, can typically deal even with those.
Routing
Harmony EFX should typically be instantiated 
as an insert effect on the track that will 
supply the source vocal. Harmony EFX can 
be instantiated on a mono or stereo track, but 
since it processes only a single channel of 
audio, if you instantiate it on a stereo track, the 
two channels will be mixed together before 
being processed.
Harmony EFX’s output can be mono or stereo. 
Unless you have some overriding reason for 
using a mono output, we highly recommend 
using it in stereo output mode, since in mono 
mode, all panning controls are disabled. This 
prevents you from positioning the harmony 
voices across the stereo soundstage and 
dramatically impacts (in a not-so-swell way) the 
overall effect.
Ef fects
The cleaner the input signal, the easier 
Harmony EFX’s task of reliably detecting its 
pitch. For this reason, effects designed to 
improve the quality of the input (e.g., de-essing, 
noise reduction, etc.) should be applied to the 
audio before it is input to Harmony EFX.
N OTE : If you will be using both 
de-essing and compression on 
your vocal track, it has been our 
experience that using the de-esser before 
the compressor provides improved results in 
Harmony EFX.
Effects like chorus or reverb can negatively 
affect Harmony EFX’s ability to reliably detect  the pitch of the input audio. As a result, you 
should not apply these effects to the input 
audio before it is routed into Harmony EFX. 
Either apply them to the Harmony EFX output 
or, if you only want them on the original audio, 
apply them to a copy of the original audio on 
another track and then mute the original input in 
Harmony EFX.
Set Input Vocal Range
In order to do its magic, Harmony EFX needs 
to figure out exactly what pitches are being 
sung. It will do its best job of pitch detection if it 
knows the general range of the vocal part. You 
can help it out by clicking on the Set Input Vocal 
Range popup menu and selecting your track’s 
vocal type. The choices are:
• 
Soprano Voice
• Alto/ Tenor Voice
• Baritone/Bass Voice
• Instrument (anything that isn’t actually a 
voice)
IM pO rTANT N OTE: We really can’t 
stress this enough. Getting this 
setting right is key in getting the best 
performance from Harmony EFX. Choosing the 
wrong Vocal  range (or just forgetting to set it at 
all) will result in compromised performance.  pay 
attention. 
						

							9
Original Input
Gain
The Gain fader controls the gain of the original 
vocal whenever the channel is not muted. 
The default Gain value is - 6dB. option (Mac)/
Control (PC)-clicking the fader returns it to that 
value. 
Solo
Engaging Solo Mode causes the original 
vocal signal to appear at the output and 
simultaneously mutes all Harmony Voice 
channels that are not also soloed. 
When the Solo button is yellow, Solo Mode in 
engaged. Click the button to toggle its state.
Mute
Engaging Mute Mode causes the original input 
signal to be muted from the output.  When the Mute button is blue, Mute Mode in 
engaged. Click the button to toggle its state.
N
OTE : If both the Input Channel’s 
Solo and Mute buttons are pressed, 
Solo takes precedence. However, if 
Solo is then disengaged, the original input will 
then be muted.
Pan
Sets the original vocal’s location in the stereo 
spectrum when its output is not muted and 
Harmony EFX is assigned to a stereo output. 
If Harmony EFX is not assigned to a stereo 
track, this control will be disabled.
Command (Mac)/Control (PC) click the control 
to reset it to its default value of 0 (center).
Tr ack i n g
As previously mentioned, for best 
performance, Harmony EFX requires a clean, 
pitched monophonic signal. 
Specifically, in order to accurately identify the 
pitch of the input, Harmony EFX requires a 
periodically repeating waveform, characteristic 
of a voice or solo instrument. The Tracking 
control determines how much variation is 
allowed in the incoming waveform for Harmony 
EFX to still consider it periodic.
If you are working with a clean, well-isolated 
solo signal you can typically set the Tracking 
control to 25 and forget it (hence, that’s the 
default value).
If, on the other hand, your signal is noisy or 
not well-isolated (as might be more common 
in a live performance situation or a track with 
headphone bleed) or you are dealing with a 
particularly breathy or guttural voice, it may 
be necessary to adjust the tracking for best 
performance. 
Interestingly (to us, anyway), we’ve discovered 
that it is almost impossible to predict, just from 
listening, what setting will work best. This led 
to our choice of labels for the extremes of the 
control’s range. 
As the labels imply, there is no hard and fast 
rule for what settings will work well with what 
kinds of input. If you’re having a problem (such  
						

							10
as octave errors or loss of pitch tracking), 
experiment until you get the best result.
A TIp: One of the things that can 
cause tracking problems is excessive 
sibilance and/or other unpitched 
articulations. De-essing (prior to compression) 
can often alleviate these problems.
I M pO rTANT N OTE: Like the Set 
Input Vocal  range setting, getting 
this setting right is critical for best 
performance. If you are experiencing problems, 
these are the controls to adjust. It may be useful to think of the four Harmony 
Voices as being similar to the voices of a four-
voice synthesizer. Each voice has independent 
controls for setting its vocal quality, pan 
and level. The method by which pitches are 
assigned to voices varies with the Harmony 
Modes selected in the Harmony Control 
section. In this section, we’ll describe the 
Harmony Voice controls. The Harmony Modes 
will be discussed later in this chapter. 
N
OTE : As mentioned below in 
the Harmony Control section, 
harmony notes will be assigned 
to the various harmony channels such that 
the lowest pitched note will be assigned to 
the highest numbered active channel and 
the progressively higher pitched notes will 
be assigned in order to the progressively 
lowered numbered active channels. For 
example, for a four note chord, the lowest 
note will be assigned to harmony channel 4, 
the next higher to channel 3, the next higher 
to channel 2 and the highest to channel 1.
Gain
Each Gain fader controls the gain of its 
Harmony Voice channel whenever that channel 
is not muted. 
In the various MIDI Harmony Modes, the Gain 
control sets the maximum gain level for its 
channel (i.e., the gain at MIDI velocity value 127). 
The default Gain value is - 6dB. o ption (Mac)/
Control (PC)-clicking a fader returns it to that 
value. 
Solo
Engaging a channel’s Solo Mode causes that 
channel’s signal to appear at the output and 
simultaneously mutes any other channels 
(Harmony Voice and/or Input Voice) that are not 
also soloed. 
When a channel’s Solo button is yellow, Solo 
Mode in engaged on that channel. Click a 
button to toggle its state.
Mute
Engaging a Harmony Voice channel’s Mute 
Mode causes that channel’s signal to be muted 
from the output. 
Harmony Voices 1– 4
Each of the four Harmony Voice channels 
are functionally identical and contain identical 
controls.  
						

							11
When a Mute button is blue, that channel’s 
Mute Mode in engaged. Click a Mute button to 
toggle its state.
NOTE : If both a channel’s Solo and 
Mute buttons are pressed, Solo takes 
precedence. However, if Solo is then 
disengaged, the channel will then be muted
Throat Length
This control uses Antares’ unique Throat 
Modeling technology to define the vocal quality 
of a Harmony Voice by actually varying the 
geometry of the channel’s model vocal tract.  
It is used in combination with the various forms 
of pitch shifting to define gender and/or vocal 
quality. 
The Throat Length control allows you to 
lengthen or shorten its Harmony Voice channel’s 
modeled throat. The range of this control is 
.75 to 1.50. Values above 1.00 represent a 
lengthening of the throat while values below 
1.00 represent a shortening of the throat. 
The values represent the percentage change 
in the throat length. For example, a value 
of 1.20 represents a 20% increase in throat 
length, while a value of 0.80 represents a 20% 
decrease in throat length. 
In addition to simply changing vocal timbre, 
increasing throat length is useful when the 
original input is female and you want the 
Harmony Voices that are being shifted down 
to sound male. Conversely, decreasing throat 
length is useful when the original input is male 
and you want the Harmony Voices that are 
being shifted up to sound female or childlike.
N OTE : While this control gives you 
the ability to radically change the 
throat length, keep in mind that 
the variation in the length of human vocal 
tracts is rarely more than about 20% in either 
direction. If you are looking for a “realistic” 
vocal characteristic, start with modest settings 
of this control. More extreme settings can 
produce dramatic results, but probably 
not what anyone would call “realistic.”
Command (Mac)/Control (PC) click the control to 
reset it to its default value of 1.00.
I M pO rTANT N OTE: This function 
is specifically for changing the 
existing quality of a voice, not 
for manual formant compensation for 
pitch shifting. Harmony EFX automatically 
applies appropriate formant correction 
when it shifts a channel’s pitch.
Interval
The Interval popup sets a Harmony Voice’s 
pitch interval relative to the original vocal in 
Fixed Interval and Smart Interval modes. 
In both modes, the range is plus- or-minus two 
octaves. In Fixed Interval mode, the popup 
displays half-steps in the range -24 to +24. In 
Smart Interval mode, the popup displays scale 
degrees in the range 16va to 16vb. 
This control is inactive in all Chord Name and 
MIDI modes.
N OTE : Harmony EFX remembers 
the most recent Interval settings for 
each of the two interval harmony 
modes. So, if you select Fixed Interval and set 
a harmony voice to +3 semitones, and then 
change to Smart Interval and set that same 
harmony voice to -5th, if you then change back 
to Fixed Interval, the Interval setting will return 
to +3 semitones.
For reference, here are the equivalent scale 
intervals for each semitone interval:
S EMITO nES = S CALE In TERVAL
1 = minor 2nd
2 = major 2nd
3 = minor 3rd
4 = major 3rd
5 = perfect 4th
6 = aug 4th/dim 5th
7 = perfect 5th
8 = aug 5th/min 6th
9 = major 6th
10 = minor 7th
11 = major 7th
12 = octave
13 = minor 9th
14 = major 9th
15 = minor 10th
16 = major 10th 
						

							12
17 = perfect 11th
18 = aug 11th
19 = perfect 12th
20 = minor 13th
21 = major 13th
22 = minor 14th (rarely used)
23 = major 14th (rarely used)
24 = 2 octaves
Pan
These controls set each harmony voice’s 
position on the stereo soundstage when 
the voice is not muted and Harmony EFX is 
assigned to a stereo output. 
If Harmony EFX is not assigned to a stereo 
track, these controls will be disabled.The control range is from -100 (panned full 
left) to 100 (panned full right). Not surprisingly, 
a value of 0 represents the center of the 
soundstage.
The default Pan values are different for each 
Harmony Voice channel as follows:
Channel 1: - 50
Channel 2: 50
Channel 3: -100
Channel 4: 100
o
ption (Mac)/Control (PC)-clicking a Pan control 
returns it to its default value. 
Humanization
The functions in the Humanize section are 
designed to allow you to add a selectable bit 
of random variation to each harmony voice, 
as would be typical of four individual singers 
singing together.
Each of these controls affect all active harmony 
voices. However, they affect each harmony 
voice individually (i.e., if you set a certain range 
of variation, the actual amount of variation will 
be a bit different for each active harmony voice).
naturalize
This control selects the amount of pitch and 
timing variation, as well as how much of the 
original input’s vibrato and pitch gestures will 
be present in the harmony voices. Unless you 
want the harmony voices to be unnaturally  perfect (which is an interesting effect), using 
this control to add the ideal (for your music) 
amount of variation will result in more realistic 
performances.
The Naturalize function’s pitch component 
applies a random variation in pitch to each 
harmony voice. The higher the Naturalize value, 
the larger the maximum amount of allowable 
variation in each voice. As mentioned above, 
the actual amount of pitch variation will be 
different (and continuously variable) for each 
active harmony voice.
The Naturalize function’s timing component 
applies a random timing variation to each 
harmony voice. The higher the Naturalize value, 
the larger the maximum amount of variation  
						

							13
in each voice. Again, the actual amount of 
variation will be different (and continuously 
variable) for each active harmony voice.
The third effect of the Naturalize function is 
to control how much of the original vocal’s 
inherent pitch variations, either vibrato or 
expressive pitch gestures like scoops and 
bends, will appear in the harmony voices. At 
a setting of 0, none of the input vocal’s pitch 
variation will be present in the harmony voices. 
NOTE : In Fixed Interval mode (only), 
the harmony voices will always 
parallel all of the original input’s 
vibrato and pitch gestures. Setting the Naturalize 
function above 0 in that case will actually 
accentuate (i.e., expand the range of) the vibrato 
and gestures. 
The range of the control is from 0 (no pitch 
or timing variation or input vibrato or pitch 
gestures) to a maximum of 100 (lots of pitch 
and timing variation and harmony voices 
duplicate the input vibrato and/or pitch gestures 
— except in Fixed Interval mode as mentioned 
above) .
o ption (Mac)/Control (PC)-clicking the knob 
returns it to its default value. 
Vibrato Controls
The following two controls are used to add a 
programmable vibrato to each Harmony Voice. 
In using these controls, it’s useful to consider 
the following principles:
•  If the original vocal does not contain its own 
natural vibrato, you can feel free to do pretty 
much whatever you want with the Vibrato 
controls. 
• If the original vocal does contain its own 
natural vibrato and the Humanize control 
has been set to allow it to be present in the 
Harmony Voices (or you are using Fixed 
Interval mode), using the Vibrato controls 
may cause interference between the natural 
and programmed vibratos, with unnatural-
sounding results. This is not to say that this 
can’t be an interesting effect, only that it is 
unlikely to sound realistic.
• If the Input Audio contains its own natural 
vibrato and the Humanize control is set to 
0 (preventing the pitch component of the 
natural vibrato from being present in the 
Harmony Voices), any loudness component 
of the natural vibrato will still be present in 
the Harmony Voices. Depending on the 
amount of loudness variation, there might still 
be interference with programmed vibrato. 
However, unless the loudness variation is 
fairly extreme, it is unlikely to be a problem.
Vibrato Rate
This control sets the rate of the vibrato (in 
Hz). The range is from 1.0 Hz to 9.0 Hz. In the 
interest of realism, the actual rate applied to 
each harmony voice will slightly (and randomly) 
vary from the set amount.
The default Vibrato Rate is 5 Hz. o ption (Mac)/
Control (PC)-clicking the knob returns it to its 
default value. 
Vibrato Pitch Amount
This control sets the depth of the vibrato’s pitch 
and amplitude modulation. The range is from 
0 (no variation) to a maximum of 100. As with 
the Vibrato Rate control above, the actual range 
applied to each harmony voice will slightly (and 
randomly) vary from the set amount.
The default Pitch Amount value is 0 (i.e., no 
vibrato).  option (Mac)/Control (PC)-clicking the 
knob returns it to that value.   
						

							14
Glide
The Glide function allows you to control the 
speed at which harmony voices transition from 
note to note during legato phrasing.
Transition Rate
This control lets you set a selectable amount 
of glide (or, for those who grew up with 
synthesizers, portamento) between overlapping 
note transitions. The control reads out in 
milliseconds and represents the time it takes to 
get halfway to the destination pitch.
NOTE : Glide only takes effect when 
successive notes overlap or butt 
up against each other. If there is 
any silence between two notes (even a few 
milliseconds worth), no glide will be applied. 
The range of the control is from 0 milliseconds 
(essentially instant transitions) to a maximum of 
1000 milliseconds .
The default Transition Rate is 125 milliseconds. 
o ption (Mac)/Control (PC)-clicking the knob 
returns it to that value. 
Harmony Control
The controls in this section allow you to choose 
the method by which Harmony EFX decides 
which harmony notes are assigned to each 
harmony voice. 
These controls, in combination with the 
Harmony Preset buttons described below,  
is where the truly unique power of Harmony 
EFX lies.
While Harmony EFX offers the usual extremes 
of automatic fixed and “smart” scale interval 
modes on one hand (very easy to use, but 
with little creative control) and the ability to 
completely specify every note of each harmony 
voice via MIDI on the other (ultimate control, 
but requiring solid expertise in note-by-note 
vocal arranging), it’s the powerful new options 
in the middle that we believe will provide you 
with innovative (and downright entertaining) 
ways of creating the harmony arrangements 
you hear in your mind. 
The desired harmony mode is selected with 
the Harmony Source popup menu. Depending 
on which mode you select, various of the other 
controls in this section will be either activated 
or disabled as necessary. Rather than discuss 
each control separately, we’ll look at all of the 
controls that function in each mode.
Here’s how they all work: