Audio Engineering Society Paper and Help?

[amirm]

Hello everyone.

Despite the popularity of ASR, I feel that a lot of people in the industry are not aware of what we have been doing. Specifically I am talking about huge library of objective review data we have created across many categories of products and some conclusions from the same. To remedy this, I thought I/we should submit a paper to Audio Engineering Society, either for convention (i.e. no peer review required), or the Journal of AES (peer reviewed with potential fees for publications). Alas, I have not honed in on the specific title and focus of the paper. The general thought is, "Learning and Conclusions from Objective Measurements/Review of 1500+ Audio Products."

The purpose of this thread is to (1) get feedback/ideas on this work and (2) get help from those of you who have written professional papers (for this or other industries) to do the actual editing, formatting, etc. of whatever we write.

My current thought is having a paper with distinct sections for speakers, headphones, amplifiers, DACs, etc. We would explain what is measured and why, followed by summary tables/graphs (e.g. SINAD, SINAD vs cost, SINAD vs country of manufacture, etc.). We would opine on why tests are selected, value and issues with each.

I think done right, this would be a very comprehensive paper with significant length. It is not typical of what AES publishes but I think that is what makes it so needed.

Assuming we move forward, the target should be AES New York Conference 2023, held October 25th through 27th. Here are the deadlines for various types of papers: https://aes2.org/contributions/aes-new-york-2023/

Papers & Express Papers:​

Peer Reviewed Papers Category 1 proposal deadline
May 31, 2023

Express Papers Category 2 proposal deadline
August 4, 2023

JAES Papers proposal deadline
August 4, 2023

Peer Reviewed Category 1 acceptances emailed
July 31, 2023

Express Papers and JAES Papers acceptances emailed
August 21, 2023

Categories 1 & 2 final manuscript deadline
September 29, 2023

I don't understand why peer reviewed papers for the conference have such quick deadline whereas the Journal papers which are also peer reviewed have a lot more time allowance.

Here are the submission requirements for J. AES papers: https://www.aes.org/journal/authors/guidelines/

Strange that they charge for 10+ pages to the tune of $100 per page! Maybe we can qualify as "Review Paper" with 20 page allowance.

Anyway, let's first hash out if want to do this and if so, exactly what we want to propose. Least hassle would be a conference paper but sure would be nice to have a Journal one.

 

[Doodski]

That would be awesome. It would lend or give you more credibility and another arrow in your quiver or golf club in your bag. I look forward to reading it!

 

[Travis]

Despite the popularity of ASR, I feel that a lot of people in the industry are now aware of what

That is supposed to be "not aware" I'm guessing?

 

[amirm]

That is supposed to be "not aware" I'm guessing?

Yes, fixed. A major reason why I need editor help. I am liable to write measurements are no good and we should all just "listen to our ears!"

 

[thecheapseats]

Yes, fixed. A major reason why I need editor help. I am liable to write measurements are no good and we should all just "listen to our ears!"

the crucible of writing always refines the ideas - but the editing in the interim can make ya' nuts...

 

[Gruesome]

Regarding the category of paper, maybe it's best to discuss the plan for the paper with the editor of the journal first and see what they recommend.

As I read it, the fees for excess pages seem simply structured to discourage research papers much longer than 10 pages and review papers much longer than 20 pages.

About the deadlines for reviewed papers for the conference issue: maybe I don't understand your question correctly, but those have to be earlier to leave time for the reviewers, including possible iterations of the paper to get it accepted.
if you just submit a regular paper for review, it will go into the next issue when it's ready, whichever issue that is.
I'm not familiar with the JAES, but does your paper have to be targeted at the conference issue (because those issues get more attention, maybe?), or could it go into a regular issue of the JAES?

 

[MaxwellsEq]

Historically the tight timescale for conferences was due to having the papers typeset and put into a (very) large and heavy book of the papers for attendees (they need something to show their boss given the very high cost of attendance). In more recent years, the papers were also put onto a CD-ROM and made available during the conference for free via an online portal for conference attendees. The committee also need to check and reject proposals for papers which are similar to other papers or have been covered a lot in the past.

I think it's a great idea. It adds gravitas to what you are doing - and JAES is likely to be strongly supportive of engineering-based objective research. You are creating a genuinely useful dataset which has a benefit for professional use of audio facilities - e.g. by demonstrating where balanced reduces noise vs where it does not (e.g. through poor implementation). You can also demonstrate the journey - e.g. what you've added over time and whether it's been beneficial and whether you've identified objective data which support a "subjective" description. You could e.g. argue that DACs and line-level control/amplification is as solved a problem as it needs to be or perhaps argue a better way of stating SINAD for products. You should also add lots of references, especially previous JAES papers, which means if you are not a member, you should join so you can scan papers which support your position.

You could also bring to the JAES' attention the benefits of "citizen" measurement. ASR has bought together people who have decades working with and measuring equipment. It's also helped build a domestic testing toolkit which is genuinely ground-breaking. And new people are starting out and supported by you and others in moving to objectively test what they buy.

A word of warning - you will be asked challenging questions if you present the paper. Sometimes the questions can be a bit... err... surprising!

 

[amirm]

A word of warning - you will be asked challenging questions if you present the paper. Sometimes the questions can be a bit... err... surprising!

Thanks for the great comments. On this, I have been beat up so much over what we are doing here that there is nothing they can ask that is worse!

 

[amirm]

About the deadlines for reviewed papers for the conference issue: maybe I don't understand your question correctly, but those have to be earlier to leave time for the reviewers, including possible iterations of the paper to get it accepted.

Let me explain more because it is confusing. Traditionally only J. AES papers were peer reviewed. Conference papers did not enjoy the same. But then they added the option of peer review for conference papers as well. What was strange to me was the deadline for peer review of the conference papers proposal is May 31 whereas peer reviewed Journal papers can go on until August 4th. Given the heavier weight of the Journal, I would have thought this would be reversed.

 

[amirm]

To get this done, we also need a research/data assistant to mine the data. For example it would be nice to compute the median frequency response and statistics of various categories of speakers and show it in an overlay graph.

I will of course be giving credit in the paper to anyone helping with these tasks. I may even sweeten the pot with some audio gear.

 

[Gruesome]

Let me explain more because it is confusing. Traditionally only J. AES papers were peer reviewed. Conference papers did not enjoy the same. But then they added the option of peer review for conference papers as well. What was strange to me was the deadline for peer review of the conference papers proposal is May 31 whereas peer reviewed Journal papers can go on until August 4th. Given the heavier weight of the Journal, I would have thought this would be reversed.

For papers going into the same issue? That is strange.

 

[Gruesome]

To get this done, we also need a research/data assistant to mine the data. For example it would be nice to compute the median frequency response and statistics of various categories of speakers and show it in an overlay graph.

I will of course be giving credit in the paper to anyone helping with these tasks. I may even sweeten the pot with some audio gear.

You need a summer student.

 

[amirm]

You need a summer student.

Indeed. Had a couple of people volunteer a few years ago but I had nothing for them to do. But now that I do, I have no volunteers!

 

[Dimifoot]

Maybe you should consider publishing/presenting the protocol in AES

Well, I am happy that you finally have decided to do so

 

[amirm]

Well, I am happy that you finally have decided to do so

I am not up for the (political) process of such standardization (think 100 meetings/committee process). But do want to introduce the abbreviated test set we are using. AES has its own protocols for DACs, ADCs, and such but as far as I am concerned, they are kitchen sink tests, mostly involving functionality than performance.

 

[BillH]

I think a review paper may be the best approach.
Typically a review paper reviews the original research while citing modern day, state of the art examples.
In your case, you would cite all the relevant research on acoustics and then present the measurements of a good speaker example. In discussing dacs and amps you would cite research on the range of human hearing, the nuances, and then cite examples of technologies that have and haven’t reached the state of transparency. You could even mention details like the experience with the buckeye connectors and the woofers whose performance is modified by a chilly garage.
Just a thought

 

[Gruesome]

To me it seems the paper is a mix of several types of publication: It is based on established/published methods of assessing audio components (DACs, amplifiers, speakers and headphones), so in that sense it has (or needs to have) a part that reviews those methods/papers; it describes - and advertises access to - a large scale technical undertaking of deciding on a set of comprehensive test procedures and testing hundreds of currently available commercial audio components (a communication perhaps?); and it describes original findings and patterns that emerge when putting these thousands of tests together (original research paper).

I still think it would be best to start by discussing with the JAES editor what format (and page limit) they would suggest.
One possible outcome might be to start with a relatively short communication (4 pages), and then expand on that by going into more depth in one or more follow-up papers, maybe separate by type of audio component.

 

[Mr. Widget]

I still think it would be best to start by discussing with the JAES editor what format (and page limit) they would suggest.
One possible outcome might be to start with a relatively short communication (4 pages), and then expand on that by going into more depth in one or more follow-up papers, maybe separate by type of audio component.

+1

Even August 4th will be here before you blink twice. I suggest giving your team and yourself more time before presenting any lengthy report on findings of the work you and ASR has been doing. Reaching out now and creating an initial paper on one of the more manageable areas of interest could work, but going full monty and being ready by August seems like a heavy lift.

 

[Short38]

Daunting task. Two thoughts: 1. Narrow the focus of the paper to dacs, amplifiers and speakers 2. Restrict the selection of items reviewed to products produced in the last two years and which are likely to be currently available.

 

[Travis]

Daunting task. Two thoughts: 1. Narrow the focus of the paper to dacs, amplifiers and speakers 2. Restrict the selection of items reviewed to products produced in the last two years and which are likely to be currently available.

Quick thought -- restricting the selection to "dacs, amplifiers and speakers" might still be way too broad. I think it would need to be electronics or speakers, not both. With AES you probably want to avoid speakers at first for various reasons. First, the CTA Standard is spinorama, so a paper on testing speakers is going to directly, or indirectly, suggest that AES adopt the CTA standard. Second, the giants of speaker measurements and testing are extensively published in AES, as well as being medal recipients so whatever paper you do on speaker testing/measurements is going to seriously require standing on the shoulders of giants to get through.*

*The "giants" of speaker testing and measurements include names like: @Floyd Toole (for significant development in the subjective and objective evaluation of audio devices); Wolfgang Klippel, John Meyer, Raymond Cooke, Stanley Lipshitz, Leo Beranek, Richard Heyser, Paul W. Klipsch, Richard Small, and Don Keele.