Brüel & Kjær's "Evolution Of Hearing Simulation - Part 2" Coming Soon
Mar 26, 2020 at 4:27 PM Thread Starter Post #1 of 16
In Brüel & Kjær's Evolution Of Hearing Simulation - Part 1 video, well-known acoustic specialist Mr. John Bareham explained the development of hearing simulation from back in the 1970’s up to 2019, and why we use ear simulators. You can watch it below:



Next week, Brüel & Kjær will be releasing Evolution Of Hearing Simulation - Part 2, in which their new measurement standard -- the Brüel & Kjær 5128 -- is explained. Based on research by Brüel & Kjær from 2003 to 2017, the 5128 provides the most human-like ear simulator to date, an important evolution beyond the simplistic geometry of current ear simulators.

We've been measuring with the 5128 since late last year here at Head-Fi HQ, and will be shifting over to the 5128 as our primary measurement fixture. The results so far have been exciting, and it will provide measurements that provide greater clarity, more closely aligning with our subjective impressions of headphones (both in-ear and around-ear/on-ear).

Please watch Part 1 above, and then Part 2 next week.

EDIT 2020-03-30 0544 EDT: Part 2 is posted immediately below. In this video (below), I narrate the research and development of the new Brüel & Kjær 5128 High-Frequency HATS, focusing on what was needed to create a human-like ear canal and inner ear. The 5128 represents the new standard for measurement -- the most human-like ear simulator available.

 
Last edited:
Mar 26, 2020 at 6:44 PM Post #2 of 16
Huh.
I wonder how much it will cost- some $30k?
I wonder if it's well dampened, since even bone, fat, and muscle can have an effect on the audio (dampening).
 
Mar 26, 2020 at 7:22 PM Post #3 of 16
...I wonder if it's well dampened, since even bone, fat, and muscle can have an effect on the audio (dampening).

Coincidentally, I was reading a paper recently that mentioned that the damping factor of the human ear canal is estimated to be three times higher than that of a corresponding brass tube.*

There's exhaustive new research behind the 5128, and it does, for example, actually mimic the transition from soft tissue to the bonier condition nearer the eardrum.

I'll be discussing the 5128 more next week.



*Hudde, H., and Engel, A. “Measuring and modeling basic properties of the human middle ear and ear canal. Part II: Ear canal, middle ear cavities, eardrum, and ossicles,” ACUSTICA - acta acustica Vol. 84 (1998) 894–913.
 
Last edited:
Mar 28, 2020 at 5:29 PM Post #5 of 16
Isn't the 5128 HATS a head and torso simulator that's been around for a few years now?...

Brüel & Kjær announced the 5128 at the 143rd AES convention in New York back in October 2017, but, if I recall correctly, it wasn't shipping more widely until last year.

...It looks like the B&K dummy head contains an IEC 60318-7 compliant coupler. Is this in any way different from IEC 60318-4?
I found this: https://www.sis.se/api/document/preview/8025984/ but it doesn't indicate any specific differences in the coupler itself.

The head and torso geometry of the 5128 complies with ITU-T Rec. P.58, IEC 60318-7 and ANSI S3.36-1985. The 5128's coupler design, however, is very different from the current standard. It's entirely new, and is part of a pinna/canal/eardrum system intended to set a new standard. (I believe the pinna/canal/eardrum assembly is available separate from the head and torso, and is called the Brüel & Kjær 4620.)

The Brüel & Kjær 5128/4620 is intended to provide more realism in audio testing, covering the full audible frequency range of 20 Hz to 20 kHz. And while it's exciting to finally see an actual defined human average response above 10 kHz, there are differences in response below that, too, versus the previous standards.
 
Mar 29, 2020 at 7:14 PM Post #6 of 16
@jude, could I make a humble, groveling request...? I know this would be a little bit of extra work for you guys, but for the foreseeable future it would be really helpful if we could see headfi measurements from both the B&K HATS kit and also from a standard 711 coupler.

Imagine this scenario. Some new coupler gets released. Initially, very few people have access to it (and maybe google searches for it even turn up empty!?). It inevitably gives results that are quite different from those of a standard 711 coupler - even below 10 kHz. We only have anecdotal evidence from the OEM that this new coupler is actually closer to that of an average human ear. Even with the best of intentions, it turns out that it actually isn't closer to what we hear than the 711 standard. And when people see measurements from this new coupler on a new headphone, they don't know whether to attribute the FR characteristics to the coupler or to the headphone. This scenario isn't all that far-fetched, is it? :wink:

If you're unconvinced, I'd love to have a chat with you about this sometime next week. We should also talk CanJam 2020. I hope I'm wrong, but I have a feeling December is still too optimistic. By sheer coincidence, I'm stuck at home for the next few weeks/months/years. You have my mobile number. Or you can pm me...!
 
Mar 30, 2020 at 1:19 AM Post #7 of 16
@jude, could I make a humble, groveling request...? I know this would be a little bit of extra work for you guys, but for the foreseeable future it would be really helpful if we could see headfi measurements from both the B&K HATS kit and also from a standard 711 coupler....

The first comparisons will be between what we've been using (GRAS RA0401) and the 5128. Later, we may take a look at the RA0045 (which we also have here). That said, many of us here have seen enough comparative measurements between RA0045 and RA0401 to know what to expect between those two -- and so we can venture to understand standard 711 in this context even before we do re-measures with an RA0045.

...Imagine this scenario. Some new coupler gets released. Initially, very few people have access to it (and maybe google searches for it even turn up empty!?). It inevitably gives results that are quite different from those of a standard 711 coupler - even below 10 kHz. We only have anecdotal evidence from the OEM that this new coupler is actually closer to that of an average human ear. Even with the best of intentions, it turns out that it actually isn't closer to what we hear than the 711 standard. And when people see measurements from this new coupler on a new headphone, they don't know whether to attribute the FR characteristics to the coupler or to the headphone. This scenario isn't all that far-fetched, is it? :wink:...

When we discuss the 5128 in more detail it'll be quite apparent why it's more human-like. There was a lot of research done between 2003 and 2017, and the evidence is far from "anecdotal." Even simple dimensional comparisons -- comparing the geometry of the current standards to the averaged human dimensions of the 5128 -- will make this clear. Additionally, there are other completely separate studies and research over the years that are corroborative that we'll also discuss. And keep in mind that the "OEM" you mentioned is Brüel & Kjær, perhaps the most respected company in its field(s) -- and from where so many of the top minds in audio test & measurement have come from over the last many decades.

When Evolution Of Hearing Simulation - Part 2 is released this week, much of this will become much clearer. I'm familiar with what's in Part 2, as I've not only seen it, I'm also the video's narrator. EDIT: See the video below.
 
Last edited:
Mar 30, 2020 at 2:52 AM Post #8 of 16
Brüel & Kjær's Evolution Of Hearing Simulation - Part 2 video was just released. In this video, I narrate Brüel & Kjær's research and development of the new 5128 High-Frequency HATS, focusing on what was needed to create a human-like ear canal and inner ear. The 5128 represents the new standard for measurement -- the most human-like ear simulator available.



For more information on High-frequency HATS, please visit: https://www.bksv.com/high-frequency-hats
 
Mar 30, 2020 at 10:19 AM Post #9 of 16
Very interesting! @jude, do you know if it's possible to buy the coupler separately, or is it tightly integrated with the new ear canal model, such that you really have to purchase the whole system? Are there any brochures or product-spec pages for the isolated coupler?

It would be enormously instructive to see comparisons against RA0045 measurements for a range of deep-fitted IEMs.
 
Mar 30, 2020 at 3:25 PM Post #10 of 16
Thanks for the explanation.
That means every older device they built for that purpose, had not the shape of the human ear canal? And they are asking 100k € for that? :smirk:
It would be the first thing coming to my mind, when building smt. like this.

This is what first came to my mind seeing the shapes of the ear canal. :relaxed:
wdfaw.jpg
 
Mar 30, 2020 at 5:08 PM Post #11 of 16
Definitely some evolutionary connection to brachiosaurus or the Loch Ness monster. Great observation @jaaibananzu :)

It looks like B&K have done a great job in extracting an average canal shape from all those MRIs. Thanks for the extra info @jude. I guess it makes sense that the canal should be an integral part of the coupler. It looks like the pinna is too? That would be very cool for earbuds and shallow-insert IEMs, but that's bound to add to the price for those that only want to measure regular IEMs. The consensus may end up being that this is the most accurate measuring system, but... Betamax tapes were superior; the Boeing 737 Max 8 had way better fuel economy. There can always be some little fly in the ointment, and here it could be price. I'd wager my mother-in-law's life that it's in the many tens of thousands of dollars, even without the torso simulator. That's bound to hamper its adoption as a new standard.

@jude - any chance of headfi producing a crinacle-like database with this new kit? You know it would get some advertising hits :wink: I would hope all OEMs would embrace the opportunity to be on such a database. Even if they don't, enough people have their own couplers these days that OEMs really can't hide their FR responses for very long.
 
Apr 3, 2020 at 11:06 AM Post #12 of 16
Very interesting! @jude, do you know if it's possible to buy the coupler separately, or is it tightly integrated with the new ear canal model, such that you really have to purchase the whole system? Are there any brochures or product-spec pages for the isolated coupler?

It would be enormously instructive to see comparisons against RA0045 measurements for a range of deep-fitted IEMs.


@csglinux, regarding your question, here is some further clarification direct from Brüel & Kjær:
Brüel & Kjær said:
The ear canal and the coupler forms an integral system with an acoustical impedance similar to that of an average human adult, hence applying just the coupler would compromise this important feature of the ear simulator as well as the acoustic loading of the device being tested. The complete ear simulator Type 4620-R-001 (right ear) and Type 4620-L-001 (left ear) is available as products that can be purchased.



Also, I'll be posting a couple of measurements from the 5128 (versus our previous reference setup) soon, with accompanying discussion.
 
Apr 20, 2020 at 1:15 PM Post #13 of 16
7323_BK_HATS_2018_Body_Image.png

On April 22, 2020, Brüel & Kjær is streaming a webinar hosted by Vince Rey titled Measuring Human Audio Perception - The Evolution Of The Head And Torse Simulator (HATS). Here's the description:

"By utilizing a new generation of high-frequency ear and mouth simulators, the High Frequency Head and Torso Simulator type 5128-C now offers unprecedented realism in audio testing, covering the full audible frequency range of 20Hz to 20kHz. This webinar will explain the important design steps to accomplish this new milestone in audio testing."

There are two sessions (click on the times below to register):
Once you pick your session, make sure to register.

The Brüel & Kjær 5128 is a substantial evolution in human hearing simulation, and is going to be our main headphone/earphone measurement fixture going forward.



On a related note, I'm almost done with the first part of a post introducing the 5128 and why we'll be using it. I'm sorry it's taking longer than expected, but there's a lot of ground to cover (and this has been a rather unusual last several weeks).
 
Apr 22, 2020 at 9:46 AM Post #14 of 16
I was curious to hear about this topic (not a professional though) but the registration form disappears after 1 second and I am redirected to a login page where I have to create an account with no clear indication of what happens next. If this is a restricted webinar (asking to create an account is not the norm for a webinar) why is it promoted on the front page of head-fi? :frowning2:
 
Apr 22, 2020 at 9:54 AM Post #15 of 16
I was curious to hear about this topic (not a professional though) but the registration form disappears after 1 second and I am redirected to a login page where I have to create an account with no clear indication of what happens next. If this is a restricted webinar (asking to create an account is not the norm for a webinar) why is it promoted on the front page of head-fi? :frowning2:

Create an account and you can register. It's free to do so.
 

Users who are viewing this thread

Back
Top