Sound Solution XAP
Sound Solution XAP
Sound Solution XAP
Features:
- Multiband Compressor-Expander-Limiter
- Center frequencies B1=170Hz, B2=1000Hz, B3=3200,B4=7200,B5=12000
- Gated AGC
- Stereo expander
- Dual band output limiter / compressor / clipper
- Bass Equalizer
- Built in preemphasis 50-75uS generator and lowpass antialias filter
What it does:
Recording or broadcasting an audio signal requires that it should be limited in its dynamic range.
In broadcasting the band must be limited in order to not interfere with the adjacent channels while
in recording studios dynamics must be limited to avoid the final media saturation (BTW, if you
analyze most CD contains severe clipping due to this effect).
This plugin realizes a complete multiband processor to “pump up” your music to make it ready to
be broadcasted or recorded into a CD.
All controls are accessible to the user for maximum flexibility, even if this could be a bit nasty for
the unexperienced people.
Functional blocks:
the functional blocks are explained in the same order that the audio signal encounters them.
STEREO EXPANDER:
it is a basic but effective stereo expander and it has two controls:
Level: the effectiveness of the stereo expander expressed in percent.
Depth: in a certain sense, is the “size” of the stereo image; large amounts can make the high
frequencies (cymbals) to loose definition.
A “stereo image monitor” allows a visual indication of the stereo image content.
AGC:
the AGC is a slow acting leveling block that keeps the input signal of the multiband processor
almost at the same level.
The “GATING” function increments the release time of the multiband processor to prevent noise
rush up during low level passages. Both AGC and the multiband compressor will not remain
“stuck” forever, but they will recover the maximum gain in about 1 minute.
The GATE THRESHOLD should be set around –30, -40dB for most music.
All compressor bands are connected to the same gating bus, except if you operate the processor in
“INDEPENDENT GATE”: in such way every band samples its input level instead of the general
input. Sincerely i cannot say to hear so much difference between the “independent” and “non-
independent” mode, but i presented this control to increase flexibility.
Finally, the “OVERDRIVE” control sets the compressor driving level; high settings results in a
louder and denser sound.
BASS EQ:
to strengthen the bass of your music you can use this small parametric equalizer.
Frequency: adjusts the frequency of the resonator. For deep bass set it around 60-65Hz, to enhance
“drum punch” it could be set around 100-150Hz.
Gain db: adjusts the effectiveness of the resonator filter. The higher the setting, the higher the
enhancing.
Remember that it cannot create “bass energy” if it is not present at all.
MULTIBAND PROCESSOR:
it is the real “heart” of this plugin and many controls and functions are available to the user.
The three processes (compressor, expander, limiter) and the five bands are arranged as a 5x3
matrix; using the “CFG” for the row and “COMPRESSOR”, “LIMITER” or “EXPANDER” for the
column you can tweak any parameter of any process.
For a more comfortable set up, the “GLOBAL” function make the parameter that is under editing to
be copied to the other bands related to the same process.
That is, if, for example, you want to set all the compressor thresolds to –6dB, simply push the
“COMPRESSOR” button and any of the “CFG” buttons of the multiband limiter.
Press the “GLOBAL” button and set the “Threshold” slider to –6dB.
Depress the “GLOBAL” button and check, pushing the other “CFG” buttons, that all bands have the
threshold set to –6dB.
The “PROCESS ACTIVE” checkbox indicates if that process (compressor, expander or limiter) is
active or not.
Compressor:
a compressor is used to decrease the dynamic range of an audio signal so that a “x” dB increase of
the input signal will result in a “x/ratio” change of the ouput.
For example, if the compressor ratio is set to “4.0” and the input signal suddenly rises of 10dB, the
output signal will rise only of “10/4.0=2.5dB”.
Below the threshold the signal is unaffected.
Expander:
an expander is the complementary process of the compressor so that the output signal is decreased
of “x*ratio” dB for an “x” dB fall of the input signal below the threshold; it is used for noise
reduction.
For example, if the expander ratio is set to “2.0” and the input signal falls 4dB below the threshold,
the ouput signal will fall “4*2.0=8dB”.
Above the threshold the signal is unaffected.
Limiter:
a limiter is an infinite ratio compressor, so for any increase of the input signal above the limiter
threshold, there is no change in the ouput signal.
As for the compressor, below the threshold the signal level is unaffected.
Ratio:
As said, it is the effectiveness of the compressor or expander.For the limiter the control is still
present but has no effect.
Threshold:
The level above or below which the compressor/expander becomes active.
Attack-release:
All these level changes are applied to the signal within some time.
The time the processor takes to adapt to an increase of the signal is called the “Attack time”, the
time it takes to recover its state after a transient is called the “Release time”.
Both affects the sound, a small “attack” time makes the processor to react very quickly to the
changes of the signal thus preventing clipping, but sounding quite unnaturally if very short times are
used. Short release times tends to create a “denser” sound, but exaggerating this effect leads to
unpleasant and fatiguing listening.
It is a wise solution to adjust the attack time according the operating band, that is using a slow
attack time (40-60mS) for the bass frequency band (band1), a medium-slow (20-30mS) time for the
low-medium band (band2) and shorter times for the high frequency bands.
Band Gain:
After the block of bandpass filters that split the audio signal into separate bands, after the
expansion/compression/limiting processes, the signal is summed again to form the broadband
signal. The “band gain” control sets the level of mixing of the various bands in order to give the
signal a sort of equalization. Adjust it according your taste.
Band Link:
It links, in percent, two bands so preventing, for 1-2 band link, an abnormal bass expansion during a
voice solo or, for the higher bands, it helps in reducing high frequency clipping.
M (Mute): as the name says, pressing it results in the muting of the related band.
It is useful to have an idea on how the bands are divided into.
Test tone: a 324Hz tone is generated to allow the alignment of the recording equipment to the
maximum output level.
Input conditioning: when pushed, it turns on and off the input conditioning filters.
Even if it is suggested to keep this feature on, you may leave it according your taste.
Bypass all: the ouput is directly connected to the input, and it is useful for comparison.
Preemphasis on: turns on the preemphasis generator. If you use this processor for FM
broadcasting, you can feed directly a stereo exciter. In this case you have to turn off the
preemphasis generator of your exciter and turn on the one into your processor.
The time costant should be selected according to the rules of your country.
Levels Monitor: it simply shows the input and output levels thus giving and idea of the processor
efficiency. The “OUT LEV” slider sets the output volume.
Load-Save Preset: the first time the plugin is loaded, it creates an ss0.dat file “DEFAULT” preset.
You can edit and save (choosing a number) up to 21 presets (0-20).
When a loaded preset is edited, an “E” (edit) will flash near the “ON AIR PROCESS” indication.
If the process is saved or another is loaded, the flashing “E” will disappear.
Some last words:
An audio processor is a nonlinear process, that is, even if the ouput signal might be slightly more
attractive respect to the input signal, it has surely more distortion respect to the input.
For this reason, always try to use the cleanest possible audio material to get the cleanest ouput
audio and remember that the correct adjustment of a multiband processor is a quite complicated
task, so, don’t be frustrated if after thirty minutes you still cannot get your sound.
Remember also that “extreme loudness” settings are well suited only for certain kinds of music
(pop, disco, rap…) while are awful for other kinds of music; so always adapt the processor to the
format of your radio.
General settings often sound well with most music formats even if they have not the same loudness.
The output sound is always a tradeoff between loudness distortion and cleaness and you must
choose the tradeoff between them according your taste.
Another last tip is that you might avoid the use of high stereo expander levels expecially with low
bitrate mp3s; mpeg puts a lot of “trash” into the stereo image so, often, enhancing the stereo image
you could enhance just “mpeg noise” (a sort of “tingling” you may hear at medium-high
frequencies).
Last recommendation is to use a good quality audio board with a “flat” output, that is no further
bass or treble equalization must be added and only in this way you may expect to have a good peak
control and a satisfactory audio quality.
The design of this processor represents my studies about DSP and C programming and the efforts of
many good people who tested this product and gave me precious indication to make it better and
better. I’m still looking for people who wants to contribute freely to this project, to make many “one
penny radios” to be on air with a satisfactory audio quality without having to spend much money
for a professional equipment.
You can contribute with suggestions about audio algorithms and or audio processing techniques
(clippers, bass enhancers, stereo expanders…) or simply with your hears to test my “frankenstein’s”
processors; i’m here to listen your ideas.
Updates in version 1.2 rev1
• Added RMS detector in multiband processor. The two bands limiter still process peak level.
• Added clipping level indication
Bugs fixed
Under development
The future
Bah!….who knows?