Headphone Correction & Spatialization

[dasdoing]

Haha, with regards to creating filters, I'm useless to you.



Yes, personally I don't think the pre-recorded demo song is a good representation, BUT it could also be a matter of headphones. There're so many variables unfortunately.

So I demoed OOYH.
unfortunatly the software is very buggy in my system. had 3 types of problems. could make it run once though (until reboot).
funny thing is how vertical and horizontal location of center varry so much with the presets. on many of them I had to adjust balance (why?).

Quad:ESL Speakers preset I guess many here will love. it a very nice deep soundstage, with a lot of boomy bass and reflections. unfortunatly a lot of decay on bass modes. whatever, this is far from what I personaly want/like. The boomy bass could actualy make those bass weak Sennheisers a little "flatter", though

AIX Recording Studio preset I loved the balanced bass response, but it has a crazy boost in the upper mids, which I later confirmed in REW

Acoustic Zen: Crescendo Speakers preset is what I liked the most. I sounds very balanced/flat. Unfortunatly before I could measure it I had to reboot and could never make the software work again (tryed for hours). This one I probably would actualy buy, if the software worked for me, and if it would work in Linux. though I would have to test it more. whatever, this is actualy the best thing I ever heard.

 

[whazzup]

So I demoed OOYH.
unfortunatly the software is very buggy in my system. had 3 types of problems. could make it run once though (until reboot).
funny thing is how vertical and horizontal location of center varry so much with the presets. on many of them I had to adjust balance (why?).

Quad:ESL Speakers preset I guess many here will love. it a very nice deep soundstage, with a lot of boomy bass and reflections. unfortunatly a lot of decay on bass modes. whatever, this is far from what I personaly want/like. The boomy bass could actualy make those bass weak Sennheisers a little "flatter", though

AIX Recording Studio preset I loved the balanced bass response, but it has a crazy boost in the upper mids, which I later confirmed in REW

Acoustic Zen: Crescendo Speakers preset is what I liked the most. I sounds very balanced/flat. Unfortunatly before I could measure it I had to reboot and could never make the software work again (tryed for hours). This one I probably would actualy buy, if the software worked for me, and if it would work in Linux. though I would have to test it more. whatever, this is actualy the best thing I ever heard.

Yes, it's really buggy, could be because of the added complexity of having its own audio driver. And in my case, I suspect that an external usb dac makes it more problematic too (as opposed to the onboard audio). And add to that my usual practice of sleeping the computer (instead of shutting it down), probably screws with the usb detection.

Yes, play with it more and see if you can eventually find a consistent way to operate the software on your computer.

 

[dasdoing]

I gave it up. I tried it hours before my last post, and hours after. there seams to be a problem with clocking, but it is the same problem with all devices. tried my interface, my TV, realtech onboard, and a virtual device. I only get cracking sound all the time. I don't know what I did diferent the time it worked, but I could never reproduce it.
Also they seam to have abandoned the project. probably because of all the problems and they couldn't solve them.

 

[phoenixdogfan]

I gave it up. I tried it hours before my last post, and hours after. there seams to be a problem with clocking, but it is the same problem with all devices. tried my interface, my TV, realtech onboard, and a virtual device. I only get cracking sound all the time. I don't know what I did diferent the time it worked, but I could never reproduce it.
Also they seam to have abandoned the project. probably because of all the problems and they couldn't solve them.

Try going to direct sound, rather than wassapi.

 

[dasdoing]

Try going to direct sound, rather than wassapi.

where? there is no such option in OOYH

 

[phoenixdogfan]

where? there is no such option in OOYH

Should be on your source (JRiver, etc).

 

[dasdoing]

I could never make OOYH work again,

instead I have done new filters from scratch.

a) the base IRs for my editing are now Shanon Pearce's mods of the original files from here https://github.com/ShanonPearce/ASH-IR-Dataset
in the past his files had very bad bass decay, but his newest version are very very clean (no clue how he did it).
but he puts a harman curve on all of those. harman curves you like or you don't, but I came to the conclusion that room IRs should be flat in the sum. why? prety simple, cause everybody allready has their prefered curves in the phones or they come harman aproximated out of the box (So the ear simulation in the IRs is not necessary). so my new filters are all almost totaly flat (+-1dB in var graphs)

b) I made some errors in convolving stages in the past. the new ones should have no convolving artifcats anymore

c) I used a mix of Denis Sbragion's DRC and REW. the new filters are therefore step response corrected. if this produces artifacts I am not sure yet. it will sure do sometimes, but most of the times the timing is just better. a topic about that with one of my new filter is here: https://www.audiosciencereview.com/...me-aligned-exess-group-delay-corrected.14489/

d) this time I made 2 filters. one is a mod of Room 28 with a lot of wall reflections like most here love. the other is a mod of Room 2, control room IRs with treated walls

*use headphones*

Room 28
https://drive.google.com/file/d/1gXM9KD8UT8nSAFxQuJ1A6lsEXyTGtjEF/view?usp=sharing
https://drive.google.com/file/d/1x1KGAIcnXS1hC3njIs3wAC3-hyUKphns/view?usp=sharing

Room 2 (for those not used to treated walls: try to hear the ambience in the recording)
https://drive.google.com/file/d/1f3rMVYJxPBu7uqBU4Jb8MkrRxMMoteN6/view?usp=sharing
https://drive.google.com/file/d/1x2yEOTW74EvoTTb1uqftXab2huEniCjM/view?usp=sharing

 

[rxp]

Hey - interesting project, wasn't aware of the repo.

I'll take a look at it in more detail - but I'm initially interested in how I perform headphone compensation. I've listened to other HRIR's recorded with Impulcifer and they sound terrible to me because it's not my own ears. One OOYH preset did sound good after a bit of EQing though. But I tried them all and that was the only one.

 

[phoenixdogfan]

Hey - interesting project, wasn't aware of the repo.

I'll take a look at it in more detail - but I'm initially interested in how I perform headphone compensation. I've listened to other HRIR's recorded with Impulcifer and they sound terrible to me because it's not my own ears. One OOYH preset did sound good after a bit of EQing though. But I tried them all and that was the only one.

You only need one.

 

[rxp]

Yep - was pretty good, but compared to the real one I made with Impulcifer it wasn't really that good.

 

[dasdoing]

Hey - interesting project, wasn't aware of the repo.

I'll take a look at it in more detail - but I'm initially interested in how I perform headphone compensation. I've listened to other HRIR's recorded with Impulcifer and they sound terrible to me because it's not my own ears. One OOYH preset did sound good after a bit of EQing though. But I tried them all and that was the only one.

which was the preset you liked?

 

[dasdoing]

I could never make OOYH work again,

instead I have done new filters from scratch.

a) the base IRs for my editing are now Shanon Pearce's mods of the original files from here https://github.com/ShanonPearce/ASH-IR-Dataset
in the past his files had very bad bass decay, but his newest version are very very clean (no clue how he did it).
but he puts a harman curve on all of those. harman curves you like or you don't, but I came to the conclusion that room IRs should be flat in the sum. why? prety simple, cause everybody allready has their prefered curves in the phones or they come harman aproximated out of the box (So the ear simulation in the IRs is not necessary). so my new filters are all almost totaly flat (+-1dB in var graphs)

b) I made some errors in convolving stages in the past. the new ones should have no convolving artifcats anymore

c) I used a mix of Denis Sbragion's DRC and REW. the new filters are therefore step response corrected. if this produces artifacts I am not sure yet. it will sure do sometimes, but most of the times the timing is just better. a topic about that with one of my new filter is here: https://www.audiosciencereview.com/...me-aligned-exess-group-delay-corrected.14489/

d) this time I made 2 filters. one is a mod of Room 28 with a lot of wall reflections like most here love. the other is a mod of Room 2, control room IRs with treated walls

*use headphones*

Room 28
https://drive.google.com/file/d/1gXM9KD8UT8nSAFxQuJ1A6lsEXyTGtjEF/view?usp=sharing
https://drive.google.com/file/d/1x1KGAIcnXS1hC3njIs3wAC3-hyUKphns/view?usp=sharing

Room 2 (for those not used to treated walls: try to hear the ambience in the recording)
https://drive.google.com/file/d/1f3rMVYJxPBu7uqBU4Jb8MkrRxMMoteN6/view?usp=sharing
https://drive.google.com/file/d/1x2yEOTW74EvoTTb1uqftXab2huEniCjM/view?usp=sharing

So it turns out using REW after DRC was a mistake as it kind of destroyed the impulse corrections DRC did.
I found out how to get a almost flat (combined) response with DRC only.
step response is cleaner/shorter now

last version:

new version:

result is a much cleaner audio, should be pretty transperent now:
EDIT: full movement

an intresting effect can be observed in popular music with a seperated center image: the center in the old version was verticaly a little too high. in the new version it is level. (too lazy to produce a track for this since nobody seams to listen them anyways lol)

 

[nhatlam96]

Is there a trial version of CanOpener Studio? Would like to try with my Focal Clear

 

[Phony846]

Is there a trial version of CanOpener Studio? Would like to try with my Focal Clear

yes, just download and register and you get a free 30-day trial.

 

[jacobacci]

Perhaps JRiver Media Centre would be a better way to go about it

Thanks @andreasmaaan for taking this initiative. I tried CanOpener as a jRiver plugin last night. Works like a breeze. jRiver has a number of other DSP plugins on board, also a parametric EQ that allows to add headphone EQ without any external processing.
I listened on my Lenovo T480s laptop with an Oriolus Mk2 IEM and I am quite taken by the soundquality of the setup. I really llike what CanOpener brings to the table.

 

[andreasmaaan]

Thanks @andreasmaaan for taking this initiative. I tried CanOpener as a jRiver plugin last night. Works like a breeze. jRiver has a number of other DSP plugins on board, also a parametric EQ that allows to add headphone EQ without any external processing.
I listened on my Lenovo T480s laptop with an Oriolus Mk2 IEM and I am quite taken by the soundquality of the setup. I really llike what CanOpener brings to the table.

Glad to hear