HeSuVi applies noise to my HRIRs
Headphone Surround Virtualizations for Equalizer APO
Brought to you by:
jak33
Menu
▾
▴
Hello
I've recently created my own HRIR with the help of the "How-To Record Impulse Responses Digitally" post provided in the wiki and it works perfectly when talking about channel correctness and stuff but whenever Audacity exports it, the file is fine. I put it in a spectrogram, no noise is to be found. However, once i put the file in the HRIRs folder and use it, noise is to be heard whenever i pause the song for the rest of the remaining room reflections. It's horrendously noticeable and ruins the feeling of pausing a track in a large reverberant room. When I put the file back to audacity for some reason I can even hear the noise, but the original copy of the file (a copy that i have never put in the HeSuVi folder) doesnt have that noise. Weirdly enough though, the spectrogram says both contain no noise at all, yet it's audibly there. This only happens to HRIRs I create myself though. Also, I've tried removing the noise with audacity but when I export it, but the same thing happens, it sounds just like the original, as long as i dont put it in hesuvi to use it as it changes it again to have noise. If anyone knows why this could be, let me know. I can provide sample audio if needed, thanks.
What excatly means "noise"? thick sound? or pop sound? or anything?
What is the total length of your hesuvi file?
white noise, 2.00 MB, 0:01.563 length
What is the source of the HRIR you recorded ?
WDR studios, SBS room. I used one angle for each according channel
So you only did some channel remapping and not recording?
Correct, the individual channels dont have any noise as ive tested each angle , its only after i map every channel to one file that the noise appears for some reason, also, if only a small region of the frequency response gets played back, theres likely close to zero noise, but pausing things like a song that has a wide frequency range, the noise is very loud
What do you mean by "only after i map every channel to one file "?
I just tested the SBS (with the included HD600 headphone filter) file with EQ-APO (without HeSuVi) and except the strong reverb as expected from the large room I don't hear noise. And when I pause the file it is silent.
Well on the official website there's only 360 angles given and not an actual 7.1 channel HRIR. Therefore, I have to use those individual recordings to select the angle i need for each channel. But now I'm curious, where do you have your SBS file from? Is it just one angle or the working HeSuVi file?
It is some time ago (more then 8 years). I used the original matlab files and the provided script (the SOFA files did not exist at that time ) to export them to 7.1 wave files.
I uploaded this files 8 years ago:
https://www.xup.in/dl,13480571/Binaurale_Wiedergabe_ueber_KH24022015.zip/
But they still need to be converted to HeSuVi. If you have python installed on your computer I can give you a simple script that does the conversion.
Last edit: reoos 2023-11-13
I do have indeed python installed, can you provide the script?
I hope it works.
It works, however, back channels are swapped with side channels but I guess I can fix that myself with audacity or by just modifying the script
You can change the channels in that line:
audiofile_new=data[:,[0,1, 8, 9, 12, 13, 4 ,3, 2 , 11, 10, 15, 14, 5 ]]
The channels are:
8 9 10 11 12 13 14 15
LS-l LS-r Rs-l Rs-r LB-l Lb-r RB-l RB-r
Last edit: reoos 2023-11-13
Last time I had "Post awaiting moderation" nothing happened for weeks.
oh so it wont show the message at all?
Yes, I only see "Post awaiting moderation".
Well I just said it works now, thanks!
By the way, do you have any other HRIRs like these? They sound really good unlike the standard hesuvi ones
Before I used my own measurement I tried some different BRIRs but beside the WDR studio measurements I only like this one (I think it is one of the two default Realizer Brirs (I did not like the BBC one)) :
https://github.com/IoSR-Surrey/IoSR_ListeningRoom_BRIRs
In the moment I don't have the 7.1 wav file anymore on my PC.
Oh okay. Do you maybe know how to convert that .sofa into the HeSuVi wavs? Also, how realistic are those measurements, does it actually sound like a real room with speakers?
No I don't know how to convert the file. The problem is that I can load the file and see the data but I don't know the coordinates that represent the different speakers.
oh okay :/
thanks anyway for the help
There is also a matlab file which is much easier to understand. I put the two wav files here:
https://1drv.ms/f/s!AnbO8YLUAW6JkA1i1SlyZTQOWcLO?e=kO2l6e
For me the center sounds shifted a little to the right but I think that I converted the file in the correct way.