Automatic resonance reduction (soothe, DSEQ...) and sound improvement (GULLFOSS; Intensity, DSM...)


Tested some auto-betterizer and auto-anti-resonance plugins with a few tracks for two tasks:

  • reducing problematic resonances that change fast on single tracks (judged by how much the result still hurts my ears)
  • generally improving the sound of single tracks and the full mix (judged completely subjectively)
  • options and flexibility (judged by how many different options there are and how helpful these seem to me)

It was just a quick and dirty thing, but I thought some people might be interested in the experiences:

GULLFOSS [multiple settings]
resonance reduction: 2
general sound improvement: 3
options: 1
notes: very subtle but nice sound improvements with soft settings

Hornet Thirty-One [Auto-EQ mode]
resonance reduction: 2
general sound improvement: 1
options: 1
notes: very inexpensive alternative to the other stuff but the Auto-EQ feature is very limited

DSM-V2 [multiple curves]
resonance reduction: 1
general sound improvement: 2
options: 2
notes: works not that great for resonance reduction imho but I like it more on groups and busses, even though because of the static curve, you have to know what kind of sound you are going for

Zynaptiq Intensity [multiple curves]
resonance reduction: 1
general sound improvement: 2
options: 2
notes: mostly strong but smooth and flattening compression especially in the mids and highs if not extreme different settings, kinda hit or miss with stronger settings, even though you can use custom curves

Oeksound Soothe 2 [multiple curves and settings]
resonance reduction: 3
general sound improvement: 2
options: 3
notes: it’s great for resonance reduction and ismilar tasks, the sharpness parameter and other options are great

TBProAudio DSEQ [multiple curves]
resonance reduction: 2
general sound improvement: 1
options: 3
notes: similar to but not as good and less sharp as soothe. Also had problems with lagging and visual feedback dropping out but might be related to the demo (soothe demo was working fine without lags)

Trackspacer [NOT USED AS INTENDED but instead sidechaining it to the same track to suppress resonances following a suggestion]
resonance reduction: 2
general sound improvement: 0
options: 1
notes: works at least a bit for soft resonance reduction, not perfect but sounds good with around 5-15%, especially when narrowing the frequencies affected

End result (reminder: I’m only testing “resonance reduction” & “general sound improvement” here):
Oeksound Soothe 2 [multiple curves and settings]: 8
TBProAudio DSEQ [multiple curves and settings]: 6
GULLFOSS [multiple settings]: 6
DSM-V2 [multiple curves]: 5
Zynaptiq Intensity [multiple curves]: 5
Hornet Thirty-One [Auto-EQ mode]: 4
Trackspacer [sidechained to the same track]: 3

Since I’m a big fan of Zynaptiq Adaptiverb and Morph, I kinda expected Intensity (and also DSM-V2) to be a good competitor in terms of general sound, but I liked what Gullfoss and Soothe did to my projects.
Trackspacer is normally intended for frequency-specific sidechaining, but works not that bad for resonance reduction. DSEQ is pretty good and not far behind soothe imho.


Gullfoss, zynaptiq unfilter and acon digital’s defilter all seem to work well for me, although I haven’t done quite the deepdive you have or really needed to. This is pretty impressive, regardless.


I downloaded a demo of gulfoss and i don’t like it.


Neutron doesn’t claim to do this, but I have gotten interesting results out of the Scultper module. I think it’s not automated EQ so much as it is some sort of multi-band compression, and it seems to want to push stuff forward rather than back. But there have been a handful of times that was just what a sound needed for me. Neutron has a bunch of other stuff going on that it can auto-mix too, but I’ve never used that.


useful info. thanks.
there is a lot of hype around this topic at the moment, its good to hear users opinions


Good suggestions - Zynaptiq Unfilter would have made much more sense in this comparison than Intensity, but it is so pricey I don’t really want to test it lol…

Downloaded the demo of Defilter and might add it too this comparison - price range is more similar to DSEQ than to soothe, so would be a nice alternative if it works well!


My experience with iZotope and especially Neutron is very mixed. I completely love RX and even the RX plugins, like some of the creative plugins and I still use the Ozone Stereo Imager and the Neutron Transient Shaper regularly, but I was a bit disappointed by Neutron 1 and the Ozone version that was out at the same time and switched mostly to other tools and haven’t upgraded anything since then I think. The automatic mixing assistant never really worked well for me and I really gave it some tries trying to improve my mixing skills. Always had to change so much it felt like it was less work and less misleading to start from scratch. But I haven’t tried the new spectral or tonal balancing stuff they added to the Neuron/Ozone plugins.


Yeah, I stay away from the AI stuff, but that spectral shaping and sculpting stuff isn’t the same. Like I said, I don’t use it much, and you probably know enough about using dynamic EQ or multiband compression that you could dial in the same results. But for me, it’s a lot easier to just use the sculpter when I’m after “that sound”. It’s very CPU intensive though, so I never use it on more than one sound in a mix.

See if there’s some demos online that show it, I think you’ll see it and figure out what it’s doing right away. For me, it’s mystifying.


a lot of people are shouting about MSpectralDynamics being able to do this kind of thing. its very difficult to test that though because it needs very careful setting up before it can do much at all. it is SO flexible though, but it takes patience to learn to use it properly.


Sculptor seems interesting, sounds good in the promo vids and seems to work similar, at least in effect, to something like DSM.
With CPU-intesive plugins like that, I keep recording to audio a lot in projects - often too early, when I should have done something else first that effects how the track I just bounced should sound in context… ^^ but it’s very helpful for sure, I couldn’t use multiple plugins like Adaptiverb, Morph or DSM on multiple tracks without bouncing.


Yeah I read about that one, too. I skipped it for now, since I’m a bit skeptical against Melda after comparing Melda MMorph against Zynaptiq Morph a lot (I bought both because I love spectral morphing) - Zynaptiq just sounds much sharper, crisper and just better for most material, no matter what settings I use in Melda MMorph. So I tend to ignore the Melda plugins most of the time, especially when looking for high quality stuff, even though I have a few that more or less do or did unique stuff such as MTransformer and MSpectralDelay.

Also tested soothe gainst DSEQ some more and now I think that with the right settings, DSEQ is really good, but the default settings for soothe work much better on most stuff, at least for the material I produce.


i never really liked morph or MMorph personally, they both dissapointed me :frowning:
i am a big fan of melda plugins. i love how much you can d owith them and how much control you have over the output.
i have never used them for anything like this though. melda has a mathematical approach, which in theory should be enough to do anything, but in practice makes them far less usable to most people.
i am not talking about their spectral vocal control though, just in general. i have no experience with either company when it comes to this kind of work. so i am not saying they compare, just sharing what i have seen people saying. FWIW most people say MSpectralDynamics does not compare to soothe (the current big name). not seen much comparison with the other options though, which is what makes your testing stand out to me <3


How did they both disappoint if I may ask? I was convinced by the concept the first time I morphed a bassline into a drum hit or a vocal into a pad or lead or whatever it did I tried first ^^

Yeah, the control options are awesome and impressive for sure! I just never got that sweet-spot spectral morphing “completely in-between two sources but still crisp/sharp/well-defined” sound out of Melda that i got out of the Zynaptiq. Definitely oculd be the case that I don’t know how to use all those options correctly but I tried a lot of changes and presets and it mostly didn’t come close.

Thanks man, I really appreciate the feedback! Looking back, I totally should have included MSpectralDynamics and also Unfilter and Defilter in the comparison, but a reason for just ignoring stuff was that I didn’t plan to post the results orginally.

Oh and I decided to get DSEQ after some more testing!


Gullfoss is amazing. At first it looked like a gimmick of how lack of controls it offered but it really delivers. I always use it on master bus now. Besides that it shows how bad your mix is. Naturally you don’t want it to do wild stuff and if you just adjust % slighty and see curve going nuts in some frequency band a good idea is to get back to stems/project file and see what’s up with that.

Heard very good things about Soothe. It has great resonance reduction. It’s very widely used, very slick interface. People constantly mentioning it.

DSEQ is suppose to be a budget option of Soothe, I took a peek and it’s also pretty good. Definitely top 3.

P.S. this is not sound/quality related but Zynaptiq products to this day still have the ugliest UI.


Completely agree with all of that. Soothe and DSEQ can also be used nicely as visual mixing helps.

Yeah, the more I was using it the more I thought that the differences to soothe are at least partly just stemming from differences in the standard threshold curve and similar stuff. Soothe has better sounding initial settings imho is a bit sharper and might handle transients better - but I’m not even completely sure about that anymore since the default settings and frequency-specific response and also the time-based settings are so different from DSEQ.

Def a really strange style…


No, this does

Also anything from Unfiltered Audio running in white mode


to be honest it has been a while since i tested either of them so i cant remember much detail about my opinions on them, but i tested MMorph after seeing morph so i could compare them. but one of the things i saw morph being sold as was another alternative way of mixing two tracks (in a DJ scenario). although that was not my intended use when i tried it i thought it would be the best test to see what they are capable of because of the complexity of the signals.
as far as i remember that is what i was dissapointed with, so its possible i could find a lot of use from them for other purposes, i can just remember the sound of them not being as smooth as i was hoping.

maybe its time i tried them again. i often forget why i disliked a plugin then end up having to try it again hahaha. i would say about 5-10% of the time i have a different opinion of it when i try it a few years later.


i saw your first example and i thought “yea fuck me thats really nasty”, but that white mess made me want to self harm. how are you supposed to even know where you have plugged things???
a good example of how it is not about colour, but about contrast.


Def try the plugins again! Lots of possibilities.

I didn’t know that it was sold as a DJ tool, but I used it as one for a quick mix a while ago, with tracks arranged in Ableton and the main Morph slider mapped to the QuNeo crossfader - I think it was one of the two mixes I did for the How Do Music releases organized by Oatbag on the old forums. It was a lot of fun to do it, but I don’t remember how the morphing sounded in that case :wink: gonna check and see if I can figure out which one it was soonish…


yea i will definitely try them again. i will use them in practical ways too, rather that just stress-test them.
to be honest i really do not like the sound of narrow spectral bands and that is probably what put me off the most. i will give them a second chance though. im feeling positive :slight_smile: