Recording Impulse Responses for Speaker Virtualization

[sander99]

2. I'm a little confused after previewing the regular tutorial on GitHub and the tutorial using Audacity.
When i see "Stereo Speaker Setup" in Github,

What is described below, is how you create a virtual 7 channel system if you have only one stereo pair of speakers to measure.
Assumed is that when looking forward the left speaker is 30 degrees left and the right speaker is 30 degrees right.
Then if you look 120 degrees left you will measure your left speaker as right side surround (it will now be 90 degrees to your right) and your right speaker as right back surround (it will now be 150 degrees to your right).
If you look 120 degrees right you will measure your left speaker as left back surround and your right speaker as left side surround.
Looking straight at one speaker you measure that speaker as your center speaker.

Put microphones in ears, put on headphones
Take heaphones off, look forward
Look 120 degrees left (left speaker should be on your right)
Look 120 degrees right (right speaker should be on your left)
Look directly at the left speaker OR...
...Look directly at the right speaker

I haven't read what is said about Audacity but I assume they just describe how you measure a complete real speaker system "at once".

Anyway, with Impulcifer you can do it both ways (and also using just 1 real speaker to create all virtual speakers one by one).

 

[sander99]

6. This is a slightly different question
If I measure my headphones with a binaural microphone in my ear, rather than virtualizing a speaker, will that give me a baseline for what my ears are actually hearing, and what my ears actually need for custom eq? I'm currently using the HD600.

Your question is formulated a bit odd. Maybe it is best if I try to explain both the speaker measurements and the headphone measurement.
The speaker measurement measures how the sound of the speakers arrives at your ear canals. That includes the frequency response of the speakers in the room, reflexions and reverberation, and the coloration by bending around your head and into your ears.
Later, to render virtual speakers, these same changes can be applied to any input signal.
If this changed signal was played back over the headphones without additional filtering then the frequency response of the headphones on your head (it is different for different people) would be "added" to the result. To avoid that, headphone compensation must be used. That means: an inverse EQ of the frequency response of the headphones on your head has to be applied.
The measurement of the headphones is needed to create this inverse EQ.

 

[musicreo]

1. i am considering the soundprofessionals ms-cb-900 microphone. I think I can connect it to my computer with a 3.5mm microphone jack, but do I really need to buy an audio interface? I have several types of DACs and AMPs, but no audio interface, so I'm wondering if I need to buy an additional one just for binaural recording.

That depends on your audio device. If your soundcard has a good microphone input there is no need for the audio interface.

3. If I understand correctly, once I've made a binaural recording with my headphones anywhere, if I want to go somewhere and extract the acoustic space of that place to my ears, I can just put the binaural microphone on my ear without a headset and record the sine sweep as it is. Is that correct?
Or do I need to bring headphones every time?

Theoretically you don't need to measure your headphones again. But you have to be sure you can adjust the mics in your ears with high precision. If the mic position changes your headphone compensation will not work anymore.

5. This question ties in with the above.
Can the SPL level at the time of recording affect the filter's preamp or overall volume?
For example, the difference between a filter derived from a sine sweep recorded at 65db and a sine sweep recorded at 85db,95db.
I was wondering if there is any difference other than the signal to noise ratio (SNR).

From my measurements I would say that it is necessary to record at high sound volume to get good results.

6. This is a slightly different question
If I measure my headphones with a binaural microphone in my ear, rather than virtualizing a speaker, will that give me a baseline for what my ears are actually hearing, and what my ears actually need for custom eq? I'm currently using the HD600.

It gives you the frequency response of the headphone and microphone (if it is flat it can be ignored) interacting with your ears at the position of the microphone.

7. If I don't use a script and just hit record in Audacity and play the sine sweep file from a USB or memory card on another unconnected speaker playback device, will I still get the exact same result I want?
There are also device configurations where the audio interface cannot be connected, so I'm curious about that as well.

The script makes the post processing much easier but it is possible to measure without it.

 

[morgin]

Came across this post on Reddit. Was wondering if this is possible and would work. I’ve tried it but asio4all gives me static noise only when using it with vbcable so can’t verify myself yet. He states we can get 9.1.6 on pc with Dolby atmos renderer app.

https://www.reddit.com/r/htpc/comme..._app&utm_name=ioscss&utm_content=2&utm_term=1

 

[musicreo]

Came across this post on Reddit. Was wondering if this is possible and would work. I’ve tried it but asio4all gives me static noise only when using it with vbcable so can’t verify myself yet. He states we can get 9.1.6 on pc with Dolby atmos renderer app.

https://www.reddit.com/r/htpc/comme..._app&utm_name=ioscss&utm_content=2&utm_term=1

Where did you get the Dolby Refrence Player? It seems you have to pay and sign up with Dolby to get it.

 

[morgin]

Where did you get the Dolby Refrence Player? It seems you have to pay and sign up with Dolby to get it.

Yeh you do but I found another place where I was able to get it. I have the .exe if you want it

 

[Lion3]

Thanks

스크립트를 사용하면 사후 처리가 훨씬 쉬워지지만 스크립트 없이도 측정이 가능합니다.

Good. Otherwise,

Should I measure using a script?

Is it not recommended to measure in REW, export to Wav file and trim in Impulcifier? I was also curious about this.

아래에 설명된 내용은 측정할 스테레오 스피커 쌍이 하나만 있는 경우 가상 7채널 시스템을 생성하는 방법입니다.
전방을 바라볼 때 왼쪽 스피커는 왼쪽으로 30도, 오른쪽 스피커는 오른쪽으로 30도 있다고 가정합니다.
그런 다음 왼쪽으로 120도를 보면 왼쪽 스피커를 오른쪽 측면 서라운드(이제 오른쪽으로 90도)로 측정하고 오른쪽 스피커를 오른쪽 후면 서라운드(이제 오른쪽으로 150도)로 측정합니다.
오른쪽으로 120도 보면 왼쪽 스피커가 왼쪽 후면 서라운드로 측정되고 오른쪽 스피커가 왼쪽 측면 서라운드로 측정됩니다.
한 스피커를 똑바로 보면 해당 스피커가 중앙 스피커로 측정됩니다.

Okay. I've only thought about 2ch, but I'll try that next time.

 

[Atriya]

This software can apparently do head tracking. Just came up on Reddit and then I read your post.

https://apl-hud.com/product/virtuoso/

Very interesting plugin, seems like a modern and more powerful Waves NX. Interestingly, no discussion on head-fi about this that I could find. Maybe I should start a thread...

 

[morgin]

Very interesting plugin, seems like a modern and more powerful Waves NX. Interestingly, no discussion on head-fi about this that I could find. Maybe I should start a thread...

I can’t get it to work. Asio4all just gives me static noise and my trial will finish soon.

 

[sander99]

I can’t get it to work. Asio4all just gives me static noise and my trial will finish soon.

Maybe you already did this correct, but just in case:
Did you notice the patchers for asio and that you need to run both the x86 patcher and the x64 patcher if you are working on a 64 bit machine?
https://give.academy/downloads/2018/03/03/ODeusASIOLinkPro/

 

[morgin]

Maybe you already did this correct, but just in case:
Did you notice the patchers for asio and that you need to run both the x86 patcher and the x64 patcher if you are working on a 64 bit machine?
https://give.academy/downloads/2018/03/03/ODeusASIOLinkPro/

Thanks I did not do this. I will give this another go. I appreciate the help

 

[Atriya]

I can’t get it to work. Asio4all just gives me static noise and my trial will finish soon.

The VST plugin worked fine for me (in Reaper). Haven't tried to use the standalone app yet.

 

[morgin]

The VST plugin worked fine for me (in Reaper). Haven't tried to use the standalone app yet.

I’m not too familiar with reaper and plugins. Never used them before and not sure if I will be able to watch movies whilst reaper is sorting out the sound with virtuoso. I’ve tried and spent hours trying to get it to work but it’s just not working for me.

Not sure why I’m chasing these because impulcifer is delivering beautifully. Also most movies don’t even use height channels or object based surround.

How did you find virtuoso does it give the proper virtual speaker effect?

 

[morgin]

For those using 7.1 can you try my BRIR and see how it sounds. As in are you getting the 7.1 in hesuvi when watching movies. I know its better when you use your own but to see how well you guys get the 7.1 surround when using someone else's.

 

[Atriya]

I’m not too familiar with reaper and plugins. Never used them before and not sure if I will be able to watch movies whilst reaper is sorting out the sound with virtuoso. I’ve tried and spent hours trying to get it to work but it’s just not working for me.

Not sure why I’m chasing these because impulcifer is delivering beautifully. Also most movies don’t even use height channels or object based surround.

How did you find virtuoso does it give the proper virtual speaker effect?

Your situation seems different from mine. I don't use this stuff for movies at all: just for converting stereo music to binaural for headphone listening. Which is why I can preprocess the music in a DAW like Reaper and don't have to worry about things like latency.

I'm comparing Virtuoso mainly with a much older software called Waves NX. It does seem to be better and also much more configurable (e.g. custom HRTFs in .sofa format) than Waves NX.