Sonarworks Headphone Calibration Software

Talk about your MIDI interfaces, microphones, keyboards...

Moderators: khz, MattKingUSA

tramp
Established Member
Posts: 1411
Joined: Mon Jul 01, 2013 8:13 am

Re: Sonarworks Headphone Calibration Software

Postby tramp » Tue Oct 17, 2017 4:10 am

Maybe this is of some use for you:
https://github.com/ShanonPearce/ASH-BRIRs

Even if it is developed for a windows EQ, the HPCFs been implemented as IR wav files and could be used with any Convolver.
On the road again.

User avatar
funkmuscle
Established Member
Posts: 2261
Joined: Mon Jun 02, 2008 2:30 pm

Re: Sonarworks Headphone Calibration Software

Postby funkmuscle » Tue Oct 17, 2017 11:12 am

tramp wrote:Maybe this is of some use for you:
https://github.com/ShanonPearce/ASH-BRIRs

Even if it is developed for a windows EQ, the HPCFs been implemented as IR wav files and could be used with any Convolver.

Hmm, thanx bro! This got me thinking. Would IR files of good mixing rooms work for headphones? I mean simulations of mixing rooms from the point of view of engineer to monitors.

User avatar
CrocoDuck
Established Member
Posts: 1052
Joined: Sat May 05, 2012 6:12 pm
Contact:

Re: Sonarworks Headphone Calibration Software

Postby CrocoDuck » Tue Oct 17, 2017 9:41 pm

funkmuscle wrote:Hmm, thanx bro! This got me thinking. Would IR files of good mixing rooms work for headphones? I mean simulations of mixing rooms from the point of view of engineer to monitors.


To cross fade? You would need 4 impulse responses:

From Right speaker to Right Engineer ear.
From Right Speaker to Left Engineer ear.
From Left Speaker to Right Engineer ear.
From Left Speaker to Left Engineer ear.

This would crossfade the sound as it was propagating in the room and you were inhabiting the engineer body. Sort of like using HRTFs really, but HRTFs recorded in a mixing room.

If your goal instead is to have a neutral sound, then definitely it will not help. Mixing rooms are designed to be neutral, but not too much. Too neutral rooms produce fatigue for the engineers. Now, you can have a very neutral IR to play with, but as long as you listen to audio through your headphone the result will be colored by your headphone frequency response.

Your headphone is a filter: it colors everything you input to it, even very neutral audio processed by a mixing room IR.

If your goal is to avoid fatigue, then a simple crossfade as you are already doing should help. It does seems that pure stereo is harder to work with for prolonged time.

If your goal is ideal neutrality I am afraid there aren't many viable ways unless you buy a probe microphone and calibrate your own headphone.

Actually...

I maybe got another idea to try to achieve some "neutralish" sound. Pick some solid sheet of something, maybe rigid thin wood. Best if you cover it with a layer of closed cell foam. Then, make a hole in the middle just big enough to insert the tip of a mic (if you have a small tipped mic, sort of like the measurement mic I posted earlier). Then, place your headphone on this flat surface, centered on the mic, and press just enough to secure a good seal: headphones operate properly only if they seal properly. Stream white noise to the headphone and capture its output with the mic. Open up a narrow band equalizer and a frequency analyzer (calf got both). Tweak the equalizer until the frequency profile of the noise you capture with the mic is as flat as you can possibly get.

This is the cheapest way to do a sort of calibration I think. It will be still unmatched at the high end when you put the headphone on your head... but maybe an improvement.

Otherwise, if you aim to do good mixes, my suggestion it to test your mixes with different headphones and speakers, check that it sounds alright in every situation. If it does, your ear is well adapted to your headphones to mix reliably: it might as well be the case.
Check my Linux audio experiments on my SoundCloud.
Browse my AUR packages.
Fancying a swim in the pond?

User avatar
funkmuscle
Established Member
Posts: 2261
Joined: Mon Jun 02, 2008 2:30 pm

Re: Sonarworks Headphone Calibration Software

Postby funkmuscle » Tue Oct 17, 2017 9:53 pm

Using the bs2b plugin to mix and when I get the Final Mix, I've been listen to it with a pair of cheap earbuds and then I go to my car and that's how I get the Final Mix now. I got something I think it could have been from the recording Revolution or something like that where the guy said that he actually mixed three songs or something like that sitting in a loud coffee shop with just headphones. He had a quality semi-open headphones and after he mixed the song he would compare it through the earbuds. He said do it at low volume and mix in mono. That's been working for me but I was just looking for something that would be a lot quicker.

User avatar
lucianodato
Established Member
Posts: 146
Joined: Sat May 15, 2010 9:00 pm

Re: Sonarworks Headphone Calibration Software

Postby lucianodato » Wed Oct 18, 2017 10:50 am

Do not over-obsess about this tools. You will get much better results if you know your headphones well. No need of fancy monitoring just test your mix in the car or some other system. Getting a flatter response for them won't save a badly recorded instrument. May I ask what headphone are you using?
Arguy (IRC)

User avatar
CrocoDuck
Established Member
Posts: 1052
Joined: Sat May 05, 2012 6:12 pm
Contact:

Re: Sonarworks Headphone Calibration Software

Postby CrocoDuck » Wed Oct 18, 2017 11:03 am

lucianodato wrote:Do not over-obsess about this tools. You will get much better results if you know your headphones well. No need of fancy monitoring just test your mix in the car or some other system. Getting a flatter response for them won't save a badly recorded instrument. May I ask what headphone are you using?


I think I agree. It is possible to get every kind of simulation/compensation/calibration you want, and it is a lot of fun to do it or think about it (for an audio tech nerd like me), but in the end:

  • You will still need to test your final mix on different devices/environment to check how well it conforms to expectations.
  • As lucianodato mentioned, if you are well adapted to your headphones and environment your ear just does the trick: it sorts of auto calibrates.

Don't get me wrong, proper mixing environments are actually important. Designing mixing rooms is hard, there are even ISO standards to do it properly, and it requires many advanced techniques. They bring many advantages, like good balance between neutral IR and low fatigue for the engineers, but these are most important if you have to work for long hours. It is possible do have very good results in less than ideal conditions.
Check my Linux audio experiments on my SoundCloud.
Browse my AUR packages.
Fancying a swim in the pond?

User avatar
funkmuscle
Established Member
Posts: 2261
Joined: Mon Jun 02, 2008 2:30 pm

Re: Sonarworks Headphone Calibration Software

Postby funkmuscle » Wed Oct 18, 2017 2:36 pm

CrocoDuck wrote:
lucianodato wrote:Do not over-obsess about this tools. You will get much better results if you know your headphones well. No need of fancy monitoring just test your mix in the car or some other system. Getting a flatter response for them won't save a badly recorded instrument. May I ask what headphone are you using?


I think I agree. It is possible to get every kind of simulation/compensation/calibration you want, and it is a lot of fun to do it or think about it (for an audio tech nerd like me), but in the end:

  • You will still need to test your final mix on different devices/environment to check how well it conforms to expectations.
  • As lucianodato mentioned, if you are well adapted to your headphones and environment your ear just does the trick: it sorts of auto calibrates.

Don't get me wrong, proper mixing environments are actually important. Designing mixing rooms is hard, there are even ISO standards to do it properly, and it requires many advanced techniques. They bring many advantages, like good balance between neutral IR and low fatigue for the engineers, but these are most important if you have to work for long hours. It is possible do have very good results in less than ideal conditions.

I do test my final mix with first some cheap earbuds, then my phone speakers to see if I've added enough harmonic distortion to the bass so it penetrates and then the car.
From what I've been told my mixes are fine lately. I just wanted a quicker way as I live on the 14th floor and trips to the underground parking to listen in the car is not fun... :D
My headphones are cheap AKG k77 semi=opened.

User avatar
CrocoDuck
Established Member
Posts: 1052
Joined: Sat May 05, 2012 6:12 pm
Contact:

Re: Sonarworks Headphone Calibration Software

Postby CrocoDuck » Wed Oct 18, 2017 5:48 pm

funkmuscle wrote:My headphones are cheap AKG k77 semi=opened.


I have no experience about this one, but I work in the headphone industry. Expensive headphones are actually often good... but I measured cheap headphones (same price range as yours) whose response compares well with headphones 10X the price.

Frequency response is not the only thing as distortion is another important factor, but that's so hard to control and predict that cheap headphones sometimes behave themselves better than expensive ones (perhaps even out of mere luck). Comfort is another important factor. There is no point in having a super good headphone if you cannot work as long as you want with it.

So, as a result, as long as you are with a comfortable headphone you are used to work with, then nothing very big is getting in the way of proper mixing.

As for this:

funkmuscle wrote:I just wanted a quicker way as I live on the 14th floor and trips to the underground parking to listen in the car is not fun...


I see what you mean, but at the same time I am not sure even ideal calibration of your headphone would ensure you can skip the testing step. Professional engineers do test their mixes on a vast variety of different conditions, even when they mix them in the best rooms. At the end, it is the real world where the stuff we produce has to sound. Mixes produced with very neutral references do tend to sound as designed in most situations. Still, it isn't something that can be taken for granted unfortunately.
Check my Linux audio experiments on my SoundCloud.
Browse my AUR packages.
Fancying a swim in the pond?

User avatar
funkmuscle
Established Member
Posts: 2261
Joined: Mon Jun 02, 2008 2:30 pm

Re: Sonarworks Headphone Calibration Software

Postby funkmuscle » Wed Oct 18, 2017 7:14 pm

Yes I am definitely sure those steps cannot be missed. Like I said it actually works after reading from the recording Revolution I think where they said after mixing with the headphones to listen to it through earbuds and even the crappy speakers out of a cell phone and the ultimate test the car.

User avatar
CrocoDuck
Established Member
Posts: 1052
Joined: Sat May 05, 2012 6:12 pm
Contact:

Re: Sonarworks Headphone Calibration Software

Postby CrocoDuck » Thu Nov 02, 2017 10:40 pm

To add some references that perhaps will expand few of my points earlier on,

CrocoDuck wrote:As I said, similar techniques to what they seem to be doing are used in pyschoacoustics experiments, where real neutrality is needed, but to work correctly a different calibration is done on each test subject.


CrocoDuck wrote:I have no experience about this one, but I work in the headphone industry. Expensive headphones are actually often good... but I measured cheap headphones (same price range as yours) whose response compares well with headphones 10X the price.


Here a couple of new research result. As you can see, these experiments use the methods I mentioned (even though from the introduction it isn't clear whether the calibration was optimized for each subject):

A Statistical Model that Predicts Listeners' Preference Ratings of In-Ear Headphones: Part 1—Listening Test Results and Acoustic Measurements

A Statistical Model that Predicts Listeners' Preference Ratings of In-Ear Headphones: Part 2—Development and Validation of the Model

One of the results states that there is a very poor correlation between headphone price and perceptual quality, as shown in this plot:

Image

Now, I haven't read the whole of the papers (my subscription expired...), and I am usually very skeptical of result published by companies (Harman in this case) rather than Universities. Usually there is some amount of marketing inside. However, all sounds pretty realistic to me. I would totally expect that.
Check my Linux audio experiments on my SoundCloud.
Browse my AUR packages.
Fancying a swim in the pond?

User avatar
CrocoDuck
Established Member
Posts: 1052
Joined: Sat May 05, 2012 6:12 pm
Contact:

Re: Sonarworks Headphone Calibration Software

Postby CrocoDuck » Sat Dec 23, 2017 8:01 am

funkmuscle wrote:so by me sending them my headphones so they can calibrate it as one of Sonarworks suggestions, would not work because I would need to be present if I understand correctly?


Hey there!

I found this presentation which is very interesting. It is from this guy, which seems to be quite an expert in the field. The presentation is mostly about spatial audio, but it contains headphone equalization as a subtopic.

It pretty much agrees with our previous findings: there is no way to calibrate headphones in a proper way by using commercially available HATS and ear modelling microphones. Also, there can be even 20 dB differences between headphone responses as measured in different people ears in the high frequency region, and most of it is due to the ear canal shape.

However, the author also says that headphone calibrated on someone else ears usually do sound fairly neutral, way more neutral than those calibrated on HATS or ear modelling mics. He also proposes a custom built HATS with ears made of silicone, out of casts of real ears, that seems to work well.

Result: if Sonarworks works for real, they are either equalizing the headphones by measuring with probe mics into a number of different human ears or they have their own custom made HATS. Personally, until I know what they do, I wouldn't consider investing or their software.

Our main conclusion is still confirmed: proper calibration should be done for each different user.

The guy has a very interesting YouTube channel in which he also explains how to build a home made probe microphone. That requires to at least have a way to measure mics frequency responses and be able to create some software (Octave or Julia will work) to compensate for the probe response when taking measurements, which should not be too hard. I honestly hope I will have time to look into it: we could create our own personal headphone calibration process out of this!
Check my Linux audio experiments on my SoundCloud.
Browse my AUR packages.
Fancying a swim in the pond?

User avatar
CrocoDuck
Established Member
Posts: 1052
Joined: Sat May 05, 2012 6:12 pm
Contact:

Re: Sonarworks Headphone Calibration Software

Postby CrocoDuck » Sat Dec 23, 2017 1:39 pm

Oh wow! There is much more to this!

I was watching this video and the guy discusses an headphone calibration method that works in a not invasive way by matching loudness. I should be much simpler than any probe mic measurement. I gotta find the way of trying it.
Check my Linux audio experiments on my SoundCloud.
Browse my AUR packages.
Fancying a swim in the pond?

User avatar
funkmuscle
Established Member
Posts: 2261
Joined: Mon Jun 02, 2008 2:30 pm

Re: Sonarworks Headphone Calibration Software

Postby funkmuscle » Sat Dec 23, 2017 1:43 pm

Dude that's amazing! If you can figure this out that would be amazing for people like myself that are stuck in the bedroom that's not setup for mixing.

User avatar
CrocoDuck
Established Member
Posts: 1052
Joined: Sat May 05, 2012 6:12 pm
Contact:

Re: Sonarworks Headphone Calibration Software

Postby CrocoDuck » Sat Dec 23, 2017 2:38 pm

funkmuscle wrote:Dude that's amazing! If you can figure this out that would be amazing for people like myself that are stuck in the bedroom that's not setup for mixing.


Seems like David already wrote a program to do it, and did a tutorial on how to use it. It looks like a Matlab thing. Not sure whether the code is available somewhere, but it looks simple in principle.

Now for the hardest part: finding time to make an enhanced cool Linux version...
Check my Linux audio experiments on my SoundCloud.
Browse my AUR packages.
Fancying a swim in the pond?

User avatar
funkmuscle
Established Member
Posts: 2261
Joined: Mon Jun 02, 2008 2:30 pm

Re: Sonarworks Headphone Calibration Software

Postby funkmuscle » Sat Dec 23, 2017 6:50 pm

CrocoDuck wrote:
funkmuscle wrote:Dude that's amazing! If you can figure this out that would be amazing for people like myself that are stuck in the bedroom that's not setup for mixing.


Seems like David already wrote a program to do it, and did a tutorial on how to use it. It looks like a Matlab thing. Not sure whether the code is available somewhere, but it looks simple in principle.

Now for the hardest part: finding time to make an enhanced cool Linux version...

Wow!! yeah time is always the factor and in my case, the knowledge. :mrgreen:


Return to “Computer Related Hardware”

Who is online

Users browsing this forum: No registered users and 1 guest