AES Headphone Technology Conference: How Researchers Go About Making the Subjective Experience Objectively Meaningful

A few of yesterdays papers on perceptual evaluations referred to the "Sound Wheel" developed by Delta's SenseLab---a Danish firm that offers an audio product testing service. I was intrigued at first simply because it was colorful and pretty. After the sessions I looked it up and found it was far more than that.

The Sound Wheel was developed to produce an objective way to perform subjective evaluations by the firm. It's of great interest to me for two reasons:

  1. The paper goes into great detail of how a series of words found in product literature, dealer advertisements, and subjective reviews are accumulated, sorted, and graded for utility. It's simply fascinating to see the scientific mind at work parsing the realm of the subjective experience.
  2. It encourages me to do a better job of writing reviews that are meaningful to as broad a range of readers as possible. I know I can be a little weird in the words I use, and I think I could do a better job by understanding and utilizing a thoughtfully developed lexicon.
  3. I'd very much like your feedback on this paper and what you think its merit and weaknesses are. You can find the paper here:

    Perceptual characteristics of audio
    The sound wheel can be used to provide an objective description of the sound.

COMMENTS
tony's picture

Standardized Terminology

Boring but has to be the starting point.

Oxford English Dictionary will have many meanings for the same word.

Describing Science with words is tricky business.

Do they have a category for Eargasm?

Tony in Michigan

PAR's picture

The wheel has antecedents for attempting to categorise subjective experience. Most notable are aroma wheels used in oenology and pioneered in California in the 1970s/80s.

Two immediate questions arise. The first is whether or not an appropriate lexicon has been chosen. For example the subject of dynamics appears to be limited to the idea of peaks and not to the ability to distinguish between small changes in low level dynamics that are so important in relation the intended musical expression.

My second question concerns the choice of words and cultural understanding. I assume that the original work was in Danish and the published wheel illustrated here is a translation. There is a question of how accurate a translation can be given the varying weights that different cultures place on different words. For example what may be considered as "bright" by Danes may, for all I know , be considered more as "shrill" by UK English speakers or "etched" by American English speakers. Just look at the very different noises in different European languages that the same common animals are thought to make.

The blind test methodology is far better than some of the vaguely bogus variety of such tests often used in the audio world. At least there is an understanding of the difference between naive and expert listeners and furthermore the test does not rely upon the use of memory.

Hi-Reality's picture

Thanks Tyll, This is very interesting and much needed in the audio industry. Perhaps a good step towards a possible perceptual evaluation standard.

A question: based on the Sound Wheel map, what is their method/procedure to actually quantify the perceived performance of an audio component/system? (e.g. has Delta developed a corresponding performance report - a sort of metrified rating chart - for cross comparison between several products?

I have to delve into the linked white paper and their website to learn more about their methods and tools.

Regards, Babak
Hi-Reality Project
www.Hi-Reality.com

AllanMarcus's picture

This is very cool and on the path to science! developing a common lexicon is the first step to then developing an objective way to measure each element of the lexicon.

The Federalist's picture

I get the idea of developing a standardized lexicon for internal product assessments.

That would help companies develop standardized polling for focus group testing. And I do appreciate them banging the blind vs. sighted drum a bit.

But once the product has made it out into the wild, the reviews are not always about accurately describing the product's sound so potential buyers can get good data ahead of purchase. Plus, the cold hard truth is that you cannot know how something sounds until you hear it for yourself. There is no standardized lexicon that can circumvent that reality.

I think everyone recognizes that highly visible reviews sell gears, but I don't think that reviews should be viewed as a wholesale extension of marketing. Rolling them into some standardized format for effectiveness and accuracy seems like a good way to make a site boring and tedious.

Sometimes the most unorthodox metaphors are the best at conveying certain things, and they can do it while being entertaining at the same time.

tony's picture

Dear Tyll,

Way back, 5 years ago + 3 months, you reviewed the Hifiman Planar headphones.

You said that you 'might' be wondering what you saw in these 4 headphones : HE-5, HE-6 and two others, in 5 Years time!

Well, I'd bet that everyone would like to know what you 'now' think.

I had a look-back to these old phones because Schiit finally got around to making an Amp ( Jot. ) able to drive these old timers. Mr.Stoddard suggests that he was planning this new design since the HE-6 era of 2012 or thereabouts. ( seems it took 4 years or more to get the Amp into Production, hmm ).

Ebay has a few HE-6 phones for sale ( $600 price range ), about 22 of them Sold, over the last 3 months. Some NEW HE-6 phones are still available in the $800 range.

The Swiss outfit Goldmund still offer a re-worked HE-6 ( Goldmund branded ) for $10,000! ( last time I checked )

My calculus reveals an $1,000 Elear Focal to be greater value for money. ( Jot.@$500 + HE-6 @$600 = $1,100 )

Am I missing something?

I hope you had a good trip to Europe ( our Ancestral Home ).

Tony in Michigan

elmura's picture

I think that is a good start. I disagree with some of the breakdown of several categories. However, with crowd sourced input, standardised definitions and breakdowns will really help make reviews & impressions comparable