DTS Sound Unbound on linux?

Yesn't. Look at HeSuVi and their process of making your own HRTF profiles. Once you'll get impulse files you can use them with pipewire, they have a preset designed specifically for HeSuVi profiles.

It's a lengthy process but the end result is roughly the same as with DTS Sound Unbound

https://sourceforge.net/projects/hesuvi/

Ugh, Sourceforge...

Take a look at easyeffects

have you tried EasyEffects?

This app is a godsend.

A bit OT: I recommend this loudness equalization preset: https://github.com/Digitalone1/EasyEffects-Presets
They plus apo files for your headphones are amazing

The audio stack for Windows and Linux are very different. As far as I'm aware it's not possible. Anecdotally I've found that these applications are pretty useless anyways, if you play any modern competitive shooter it will have its own HRTF audio processing engine that works better anyways. I know CS2 and Valorant do. It's not possible to get more accurate audio unless you know the actual position of the enemies which these programs do not.

Other than that good headphones with good imaging really helps. I.e. not typical gaming headphones.

No, that's a windows program and running it in a VM would probably cause unacceptable latency. You can use PipeWire or JACK to put real time audio filters inline though.

Or passthrough the VM a USB-DAC, no audio in Linux then though

Not DTS but I tried this pipewire filter some time ago and it works pretty well.

No.

But why would you want to?

All "virtual surround" solutions for headphones are absolute crap. You have two ears. The headphones have two speakers. Any modern game is already processing the 3D position of sounds and creating the appropriate stereo signal.

All any additional processing does is make it sound worse. At best these programs are just an audio equaliser that makes footsteps louder, that's not some fancy processing, and is arguably cheating.

You should read up on head related transfer functions. Virtual surround is much more than just "an audio equaliser that makes footsteps louder".
- No fucking shit.
  
  That's not what I'm saying. These programs don't do virtual surround. If they do, they do it badly.
  
  Game engines DO use virtual surround, and do so much better, using it to produce the stereo signal directly.
  
  These programs, at worst, pretend to accept a 5.1 or 7.1 audio source, and then downmix it, badly, to stereo. They might slap on a shitty room effect for good measure.
  
  At best, they just take the stereo audio from the game, which is either already in virtual surround, or a stereo stream without positional data, and muck about with the levels a bit.
  
  That's all they can do. Because what they claim to do, which is add surround when it isn't there, or improve it when it's already there, isn't a thing.
  
  These programs don't "enable" virtual surround. The real thing is already in every modern game engine ever.
  
  And if a game engine isn't already doing HRTF, some random program isn't gonna add it by "processing" the stereo signal the game IS producing.
  
  To do that, the game engine would need to be providing all the sounds and source locations, walls that the sound is to interact with, etc. But guess what, if it supports doing all that, IT'S PROBABLY ALREADY DOING HRTF.
Game's well have great cinematic sound but doesn't mean it's the best choice. As example CS,Rainbow Six Siege etc you would prefer hearing clear steps. So someone else with "bad" sound could hear you walking while you couldn't hear him.
- That's what I said.
  
  If someone else turns up footstep audio with additional software, you have to do it too.
  
  But just use a damn equaliser then. Not some "surround gaming audio" processor that at worst does way more to hurt audio quality than help it.
  
  These programs don't help you hear other players by using some magic 3d audio. They literally just make footsteps louder, because the audio is already 3d.
  
  You can also get the same result by just lowering all the audio sliders in the game, except SFX, and then using a louder volume.
  
  These "audio processing" programs are fake garbage.