Impulcifer, Copy speaker sounds to headphones!

[PureLIN]

how to use it.

To use it , you need 12 or 60 captured impulse file (not like hesuvi, 1 file only contains 1 postion to left ear and right ear), a sensor for get position data (I recommend valve index, sensor very precise and has good over ear spakers).
And install my homebrew EqualizerAPO (Just install the offical version and replace the EqualizerAPO.dll/VoicemeeterClient.exe, you need shut audiodg.exe if can't replace dll)
Add the example config and it should ready to go.

If you don't have sensor, here is a python script could make sound moving around (it will simulate head yaw and roll)

 

[kthulhutu]

@jaakkopasanen I'm interested in giving this a try with a friend.

Just two questions.

1. Would this work with bluetooth headphones?
2. Is it possible to just do L+R speaker measurements if you don't care about surround simulation?

 

[jaakkopasanen]

@jaakkopasanen I'm interested in giving this a try with a friend.

Just two questions.

1. Would this work with bluetooth headphones?
2. Is it possible to just do L+R speaker measurements if you don't care about surround simulation?

Yes and yes. Bluetooth has no impact on this. In the measurements guide the first thing to do is a stereo setup before getting into more complex surrround stuff.

 

[oivavoi]

Just discovered this. Looks VERY interesting.

A few quick and dumb noob questions: does it run on macs, or only on windows or Linux?

How does the sound playback or "output" actually function? Is there a separate playing app or something like that, or does it change the audio output of the pc as such for all audio?

Is there any way of using the calculated algorithm on more mobile devices than a pc, such as headphone amps with dsp equalizers etc?

Or is there a general explanatory article somewhere where all these things are explained? Tagging @jaakkopasanen

 

[jaakkopasanen]

Just discovered this. Looks VERY interesting.

A few quick and dumb noob questions: does it run on macs, or only on windows or Linux?

How does the sound playback or "output" actually function? Is there a separate playing app or something like that, or does it change the audio output of the pc as such for all audio?

Is there any way of using the calculated algorithm on more mobile devices than a pc, such as headphone amps with dsp equalizers etc?

Or is there a general explanatory article somewhere where all these things are explained? Tagging @jaakkopasanen

Impulcifer does only the measurements, not real-time processing. You need a different app for that. On Windows, EqualizerAPO and especially with help of HeSuVi can do that for you. On Linux there's module-virtual-surround-sink for PulseAudio. I don't know what exists on Mac but I imagine some VST plugin with a host would do the work. PC is required as basically none of the independent devices have the processing power to do this many convolution operations in real-time, with the exception of Smyth Realizer and few other dedicated products. The Impulcifer repo in Github and the Wiki are where the documentation lives.

 

[oivavoi]

Impulcifer does only the measurements, not real-time processing. You need a different app for that. On Windows, EqualizerAPO and especially with help of HeSuVi can do that for you. On Linux there's module-virtual-surround-sink for PulseAudio. I don't know what exists on Mac but I imagine some VST plugin with a host would do the work. PC is required as basically none of the independent devices have the processing power to do this many convolution operations in real-time, with the exception of Smyth Realizer and few other dedicated products. The Impulcifer repo in Github and the Wiki are where the documentation lives.

Thanks a lot!

 

[q2klepto]

Just found out about hesuvi and hrir.. This is interesting as I want to use headphones for night listening

Is it possible to record hrir using minidsp ears?

I already own that rig and would be nice not to buy anymore equipment

Also, unless I'm misunderstanding.. Does this process mimic the actual sound signature of your speakers to play them back on your headphones

For eg I have nice speakers in the LS50 Metas and a 13.1.4 home theater setup... But my headphones are of much higher quality with the Hifiman Susvara

I would like the quality of the susvara but with the positioning / imaging of my home theater

 

[jaakkopasanen]

Just found out about hesuvi and hrir.. This is interesting as I want to use headphones for night listening

Is it possible to record hrir using minidsp ears?

I already own that rig and would be nice not to buy anymore equipment

Also, unless I'm misunderstanding.. Does this process mimic the actual sound signature of your speakers to play them back on your headphones

For eg I have nice speakers in the LS50 Metas and a 13.1.4 home theater setup... But my headphones are of much higher quality with the Hifiman Susvara

I would like the quality of the susvara but with the positioning / imaging of my home theater

You could measure HRIR with MiniDSP Ears but I'm not sure why you would. The idea is to measure with in-ear microphones to capture your own personal head related transfer fuction because that's what your brains are used to and nothing else will sound convincing. The HRIR captures speakers, room, your head and ears and includes headphone compensation. This means it will also include any flaws in your speakers and room acoustics.

 

[q2klepto]

You could measure HRIR with MiniDSP Ears but I'm not sure why you would. The idea is to measure with in-ear microphones to capture your own personal head related transfer fuction because that's what your brains are used to and nothing else will sound convincing. The HRIR captures speakers, room, your head and ears and includes headphone compensation. This means it will also include any flaws in your speakers and room acoustics.

I already had the EARS on hand and wanted to try it.

Surprisingly works really well.. AB testing is very very close and can only tell on bass heavy material

I may get binaural mics and an interface but it works well enough I think

 

[Smaestro]

Continuing the discussion from this post

Master Complaint Thread About Headphone Measurements

I think this long write-up is beside the point. I understand Harman is a well-researched preference, I have no qualms about that. The point I was responding to initially was this statement by Amir: The first bolded part is about that if the producing side of music has disagreeing opinions...

www.audiosciencereview.com

With @Mr. Haelscheir

Fiest of all, sorry for the late reponse. I had a lot of things happening and I couldn't get back to you sooner.

In the meantime I've tried numerous binauralisers:
- Hesuvi with all the supplied HRTFs
- Equaliser APO with the two crossover options (Jan Meier and ...)
- APL Virtuoso
- My own homebrew plugin stack with EQ'd and Delay crossover.

So this is without any personalised earfish HRTF.

Out of these, APL Virtuoso is the clear winner of giving me the illusion of sitting in a room listening to speakers. It blew my mind to be honest, as if my headphones disappeared. It has gained a permanent place in my monitoring chain, and I use it now 90% of the time mixing.

On the bottom rank are the HRTFs in the Hesuvi pack. I went through all of them, and none of them sound even remotely good to me, in any way. Either weird and unpleasing EQ and phasing, or no noticeable virtualisation effect.

The Jan Meier algorithm in Equaliser APO is alright. A bit more convincing than my homebrew chain, but also a bit more unbalanced in the EQ. Both are servicable but not great.

Have you tried APL Virtuoso?

 

[Mr. Haelscheir]

Continuing the discussion from this post

Master Complaint Thread About Headphone Measurements

I think this long write-up is beside the point. I understand Harman is a well-researched preference, I have no qualms about that. The point I was responding to initially was this statement by Amir: The first bolded part is about that if the producing side of music has disagreeing opinions...

www.audiosciencereview.com

With @Mr. Haelscheir

Fiest of all, sorry for the late reponse. I had a lot of things happening and I couldn't get back to you sooner.

In the meantime I've tried numerous binauralisers:
- Hesuvi with all the supplied HRTFs
- Equaliser APO with the two crossover options (Jan Meier and ...)
- APL Virtuoso
- My own homebrew plugin stack with EQ'd and Delay crossover.

So this is without any personalised earfish HRTF.

Out of these, APL Virtuoso is the clear winner of giving me the illusion of sitting in a room listening to speakers. It blew my mind to be honest, as if my headphones disappeared. It has gained a permanent place in my monitoring chain, and I use it now 90% of the time mixing.

On the bottom rank are the HRTFs in the Hesuvi pack. I went through all of them, and none of them sound even remotely good to me, in any way. Either weird and unpleasing EQ and phasing, or no noticeable virtualisation effect.

The Jan Meier algorithm in Equaliser APO is alright. A bit more convincing than my homebrew chain, but also a bit more unbalanced in the EQ. Both are servicable but not great.

Have you tried APL Virtuoso?

Funny enough, per https://www.head-fi.org/threads/rec...-virtualization.890719/page-121#post-18027627 (post #1,812), APL Virtuoso at least as run within Reaper was actually one of the worst for my ears at least when simulating rotations considering that I couldn't get my head-tracking to work with it. Oh; it looks like I hadn't described there the issue that while it could yield nicer room simulation spatial qualities (but did not allow completely disabling room reflections to simulate the clarity of anechoic listening), even with my personal HRTF, I was hearing much more colouration of the tonality when using the rotation sliders than with the other binaural decoders. Otherwise, from my measurements of its transfer functions in that link, it like others except for CroPaC exhibited tonal issues with how the phase implementation affects the combined response from both channels and hence the phantom center.

The second part of https://www.head-fi.org/threads/anyone-into-crossfeed.961533/post-18068536 (post #78) covers how to set up CroPaC for non-head-tracked usage of the default HRTF. AmbiRoomSim gives nice flexibility for room simulation and virtual speaker placement plus the option to completely disable room reflections which I find gives the clearest-sounding results though some gentle reflection simulation can help with immersion.

 

[Smaestro]

You're right in that Virtuoso isn't suitable for anechoic simulation. For my use case I specifically want a room sound, so that is at least one half of why I like it.

I see that CRoPaC performs very well in your measurements. I'll follow your guide and give it a try when I get back from vacation in two weeks.