Auditory Scene Analysis

[Phelonious Ponk]

Tim, I just updated/changed my previous reply to your post - it was late last night when I was replying & I didn't have the energy for a full reply. Yes, I understand your point & I believe I've answered it in my update?

Here?

"Sure, the missing fundamental is an easy example so we know what to measure to ensure that the missing fundamental is correctly recreated but there are lots of other processing tricks used that we don't know about so we don't know what/where to measure & to what level. So let's say that we didn't know the missing fundamental trick & how it works & we measured our waveform. Let's say that we find a slight discrepancy in the 5th harmonic but it is considered of no consequence as it is considered below audibility, yet it effects the perception of the fundamental, skews it in some audible way. Not knowing where to focus we see nothing wrong with the measured audio signal - all measurements are fine in the area above "audibility" so we conclude it's a delusion, right? This is where I'm saying we are with current measurements - we don't know where to look & we are using old audibility thresholds which need to be re-evaluated."

Is that speculative or is there something in this research that indicates that inaudible frequencies are triggering perceptual responses? I'm pretty sure that in music, if the 5th harmonic is inaudible and you're still getting the implied fundamental, the answer is much simpler: Enough harmonics are audible that the fundamental is implied without the missing 5th.

Tim

 

[John Kenny]

Here?

"Sure, the missing fundamental is an easy example so we know what to measure to ensure that the missing fundamental is correctly recreated but there are lots of other processing tricks used that we don't know about so we don't know what/where to measure & to what level. So let's say that we didn't know the missing fundamental trick & how it works & we measured our waveform. Let's say that we find a slight discrepancy in the 5th harmonic but it is considered of no consequence as it is considered below audibility, yet it effects the perception of the fundamental, skews it in some audible way. Not knowing where to focus we see nothing wrong with the measured audio signal - all measurements are fine in the area above "audibility" so we conclude it's a delusion, right? This is where I'm saying we are with current measurements - we don't know where to look & we are using old audibility thresholds which need to be re-evaluated."

Is that speculative or is there something in this research that indicates that inaudible frequencies are triggering perceptual responses? I'm pretty sure that in music, if the 5th harmonic is inaudible and you're still getting the implied fundamental, the answer is much simpler: Enough harmonics are audible that the fundamental is implied without the missing 5th.

Tim

You're kinda stuck on this "inaudible" idea. So let's tease it out - first there's a difference between whether the eardrum vibrates or not - if it doesn't vibrate then agreed, there's no electrical signal generated, right? This is not the same as audibility - audibility is whether we perceive this vibration, whether it registers as a conscious auditory entity. This audibility has been established using test tones, in the main. Test tones are not a valid test of the brains auditory processing because there's nothing to process.

The phenomena of hearing below the noise floor is to do with processing power, not with the ear.

Do you get what I'm saying? Give the brain full music & there's lots of signals available for the pattern matching processes that we are excellent at.

 

[Phelonious Ponk]

You're kinda stuck on this "inaudible" idea. So let's tease it out - first there's a difference between whether the eardrum vibrates or not - if it doesn't vibrate then agreed, there's no electrical signal generated, right? This is not the same as audibility - audibility is whether we perceive this vibration, whether it registers as a conscious auditory entity. This audibility has been established using test tones, in the main. Test tones are not a valid test of the brains auditory processing because there's nothing to process.

The phenomena of hearing below the noise floor is to do with processing power, not with the ear.

Do you get what I'm saying? Give the brain full music & there's lots of signals available for the pattern matching processes that we are excellent at.

Yeah, I think audibility is pretty fundamental to audio perception. Is there anything in this research that indicates that frequencies outside of the audible range are generating signals to the brain? That would be interesting. Please quote it here. But if not, again, you're just speculating, and this body of research is not relevant to the point you appear to be trying to make, the point that you always try to make - that audio performance is impacted not only by what falls outside of what is measurable, but what falls outside of audible.

Tim

 

[fas42]

John, you're familar with the ASA ideas to some degree - does the concept of the ear's AGC or whatever it's called come into it strongly? By this I mean that the ear automatically adjusts to the volume/intensity of the sound, and is able to "absorb" the peaks while still hearing through to the much lower level detail - the mind is countering the simplistic view of "masking" that is usually bandied about.

I mention this, because in my listening to the system when it's working to a high order this is what happens - I can listen very closely to a driver, the sound level is quite intense in the peaks, but I can "hear past" those peaks and easily discern the much softer activity occurring between those maximums - and if that quieter sound is lacking in quality it is quite obvious in this situation.

 

[John Kenny]

Yeah, I think audibility is pretty fundamental to audio perception. Is there anything in this research that indicates that frequencies outside of the audible range are generating signals to the brain? That would be interesting. Please quote it here. But if not, again, you're just speculating, and this body of research is not relevant to the point you appear to be trying to make, the point that you always try to make - that audio performance is impacted not only by what falls outside of what is measurable, but what falls outside of audible.

Tim

Tim, again your looking at it as in frequencies considered outside of audibility - I'm talking about amplitudes below which it's considered inaudible - two very different things

 

[John Kenny]

John, you're familar with the the ASA ideas to some degree - does the concept of the ear's AGC or whatever it's called come into it strongly? By this I mean that the ear automatically adjusts to the volume/intensity of the sound, and is able to "absorb" the peaks while still hearing through to the much lower level detail - the mind is countering the simplistic view of "masking" that is usually bandied about.

Yes, there's a number of concepts that come into play that I know of - there's feedback to the physical mechanism of hearing in the ears which adjusts this AGC & allows us to be able to have this amazing dynamic range of hearing. It is also controlled by attention & in a more general sense by top down processing. What this means is that there are two models of hearing bottom-up & top-down. In bottom-up, the signals determine the auditory objects, in top-down we compare the signals with our stored auditory models & predict which model it belongs to thus anticipating what other signals we should hear which belong to this model - this causes feedback to the hearing mechanism to focus & filter for these signals. A continuous comparison of signal to model is happening & the model changed if it doesn't fit the signals.

I mention this, because in my listening to the system when it's working to a high order this is what happens - I can listen very closely to a driver, the sound level is quite intense in the peaks, but I can "hear past" those peaks and easily discern the much softer activity occurring between those maximums - and if that quieter sound is lacking in quality it is quite obvious in this situation.

I suspect that this is what I mentioned about visual perception - our processing isn't actually dealing with "objects", it's dealing with "features" & we can combine these features in a a number of ways to focus on certain aspects of the "scene". Perceiving the "full scene" is an illusion - we never do, we are always sensing only a part of the scene - the part we are focussed on (I'm typing this & if I pay attention, I'm aware that everything else is out of focus except the screen & keyboard). It appears that we are sensing the whole scene because we can quickly change our focus to any aspect of the scene, we like - just like a computer appears to be doing many thing at once but it isn't.

 

[fas42]

I'm reminded here of our visual ability to handle HDR, high dynamic range, situations - say, looking at a bright, cloud filled sky with a tree throwing a shadow over the garden at the same time. We "think" we see the clouds and the garden at the same time, but taking a photo "proves" we don't: the contrast is immense, it's impossible to register the intensity gradation in one "gulp" - the mind flicks rapidly up and down, switching between the two "scenes" and strongly compresses them - yes, classic dynamic compression - so it appears as one to our mind - then we get annoyed when the camera can't capture it!!

 

[John Kenny]

Frank, if we had some real experts in psychoacoustics here we would all learn so much but I haven't discovered any (despite Dynamic claim that he knows more in his little finger about psychoacoustics than I ever will ). I've also not found any forums of such experts where questions of this nature could be asked. So we are ploughing unturned land & the furrows are difficult to make - particularly as perception & ASA is a very complex area - not helped by the fact that we are dealing with a playback illusion generated by technology (not to mention the constant cries of "waffle" & bickering coming from "the other side"). So understanding & dealing with all these facets of the issue is quite difficult & I don't think any one person can do it - it requires experts in each of these areas to co-operate & share knowledge in an open-minded collaborative way. Yea, like that's gonna happen

 

[John Kenny]

I'm reminded here of our visual ability to handle HDR, high dynamic range, situations - say, looking at a bright, cloud filled sky with a tree throwing a shadow over the garden at the same time. We "think" we see the clouds and the garden at the same time, but taking a photo "proves" we don't: the contrast is immense, it's impossible to register the intensity gradation in one "gulp" - the mind flicks rapidly up and down, switching between the two "scenes" and strongly compresses them - yes, classic dynamic compression - so it appears as one to our mind - then we get annoyed when the camera can't capture it!!

Indeed, the first step in the understanding is to realise that all perceptions are a construct - we are not cameras viewing the world, we are computers processing signals - we are nor audio recorders sampling the world, we are computers processing the signals, etc

The processing is where it's all at & where we need to focus if we are to break through our current limited understanding.

I can understand Tim's points & I'm trying to address them in the best way I know - I'm sure an expert would do a much better job - Dynamix where are you

PS, And I'm sure I will get things wrong - I don't have a compulsion to always be right.

 

[John Kenny]

Yeah, I think audibility is pretty fundamental to audio perception. Is there anything in this research that indicates that frequencies outside of the audible range are generating signals to the brain? That would be interesting. Please quote it here. But if not, again, you're just speculating, and this body of research is not relevant to the point you appear to be trying to make, the point that you always try to make - that audio performance is impacted not only by what falls outside of what is measurable, but what falls outside of audible.

Tim

Tim, to answer your question directly, I believe there are reports using fMRI showing that signals at frequencies outside of audibility create electrical activity in the brain but I can't find them at the moment.

 

[Phelonious Ponk]

Tim, again your looking at it as in frequencies considered outside of audibility - I'm talking about amplitudes below which it's considered inaudible - two very different things

John, again, is there anything in this research indicating that sounds outside of the accepted audible range cause activity in the brain? If yes, please quote it here. If no, then this research is not relevant to the point you're trying to make. That doesn't mean it's not interesting or worth discussing, but it does mean that it doesn't support your argument.

Tim

 

[John Kenny]

The next step in understanding is that the signals that we have to work on are not enough to fully determine an exact object - in other words we usually don't have enough visual data to make a full determination of what we are seeing; we don't have enough auditory data to make a full determination of what we are hearing, etc. As I said, it's like sitting at the side of a swimming pool & using the waves lapping into the corner to determine how many people are in the pool, where they are in the pool, how are they moving, what's the size of the pool, etc.

But we can't exist in a state of unknowing- it causes anxiety as we have to deal with the world in real-time. This means we use all sorts of techniques to alleviate this state of anxiety - so all our senses are multi-modal - hearing uses sight (among other things) & sight uses hearing (among other things).

The immediate implication of this that springs to mind is that why, when we do blind testing, we often second guess ourselves

 

[John Kenny]

John, again, is there anything in this research indicating that sounds outside of the accepted audible range cause activity in the brain? If yes, please quote it here. If no, then this research is not relevant to the point you're trying to make. That doesn't mean it's not interesting or worth discussing, but it does mean that it doesn't support your argument.

Tim

Tim, cut me a bit of slack, please - I'm trying to answer you as best I can but you can also do some research yourself to answer your own questions if you are genuinely interested - it's not battle of who's right & who's wrong here. I did indicate that there were reports of this in the post just above your post asking the question!
Here's a start for you "Can you actually hear 'inaudible' sound?"
"They have found out that humans can hear sounds lower than had previously been assumed. And the mechanisms of sound perception are much more complex than previously thought."

And here
Frequencies of Inaudible High-Frequency Sounds Differentially Affect Brain Activity: Positive and Negative Hypersonic Effects
"The hypersonic effect is a phenomenon in which sounds containing significant quantities of non-stationary high-frequency components (HFCs) above the human audible range (max. 20 kHz) activate the midbrain and diencephalon and evoke various physiological, psychological and behavioral responses. Yet important issues remain unverified, especially the relationship existing between the frequency of HFCs and the emergence of the hypersonic effect."

And if you want to go outside psychoacoustics & look at some other issues "
Sound from ultrasound"

 

[Phelonious Ponk]

Tim, cut me a bit of slack, please - I'm trying to answer you as best I can but you can also do some research yourself to answer your own questions if you are genuinely interested - it's not battle of who's right & who's wrong here. I did indicate that there were reports of this in the post just above your post asking the question!
Here's a start for you "Can you actually hear 'inaudible' sound?"
"They have found out that humans can hear sounds lower than had previously been assumed. And the mechanisms of sound perception are much more complex than previously thought."

And here
Frequencies of Inaudible High-Frequency Sounds Differentially Affect Brain Activity: Positive and Negative Hypersonic Effects
"The hypersonic effect is a phenomenon in which sounds containing significant quantities of non-stationary high-frequency components (HFCs) above the human audible range (max. 20 kHz) activate the midbrain and diencephalon and evoke various physiological, psychological and behavioral responses. Yet important issues remain unverified, especially the relationship existing between the frequency of HFCs and the emergence of the hypersonic effect."

And if you want to go outside psychoacoustics & look at some other issues "
Sound from ultrasound"

Thank you.

Tim

 

[John Kenny]

Tim, I answered your post about "inaudible" sounds causing brain signals - did you read the references I gave?
Now you want to argue another point - what are you trying to prove here - that I'm wrong? I may well be but I was hoping that this thread might not descend into a debating contest

No, that's the next step for something, perhaps, but not for supporting your point.

You keep saying that but remind me what point it is?

If that theory is accurate, it says is that perception fills in blanks not in the original information that reaches the ears. What you're supposing is that there is extra information simulating the ears that is outside of the accepted audible range; extra, unmeasurable data, that the ears transmit to the brain, and the brain uses to fill in those blanks. If the researchers were trying to establish that, step one would be measuring brain activity when "listening" to subsonic and supersonic sounds. These researchers are not doing that because they're not trying to make your point, which is probably why you've been unable to connected this research to your hypothesis in any way.

Tim

What? Did you read the papers I referenced (you are some speed reader, if you did). What is it that the papers do not support?

Ah, right, I see you are catching my posts now!!

 

[Phelonious Ponk]

Tim, I answered your post about "inaudible" sounds causing brain signals - did you read the references I gave?
Now you want to argue another point - what are you trying to prove here - that I'm wrong? I may well be but I was hoping that this thread might not descend into a debating contest
You keep saying that but remind me what point it is? What? Did you read the papers I referenced (you are some speed reader, if you did). What is it that the papers do not support?

Ah, right, I see you are catching my posts now!!

No, I didn't read it and doubt you did until I pushed. Otherwise, you would have started off by presenting the bit of information in there that supports your POV.

Tim

 

[Phelonious Ponk]

No, I didn't read it and doubt you did until I pushed. Otherwise, you would have started off by presenting the bit of information in there that supports your POV. In spite of that support, as a DAC designer, I'd still recommend that you start with the ability to measure and analyze your products. First things first, unless you're just looking for ideas to justify your lack of the proper equipment to do so. After you have your products working, and measuring at absolute state of the art within the audible spectrum, you might start looking at the possibility of improvements beyond audibility having a positive impact on perception.

Tim

 

[John Kenny]

No, I didn't read it and doubt you did until I pushed. Otherwise, you would have started off by presenting the bit of information in there that supports your POV.

Tim

Jeez, Tim, do you think I have instant recall of everything I have read & can lay my fingers on it - I knew I had read reports of fMRI showing that "inaudible" signals created brain waves & I said so.

Christ you are some cantankerous ol' bugger, aren't you
Not to worry, I'm one too but let's keep this civil, OK?

 

[Phelonious Ponk]

Jeez, Tim, do you think I have instant recall of everything I have read & can lay my fingers on it - I knew I had read reports of fMRI showing that "inaudible" signals created brain waves & I said so.

Christ you are some cantankerous ol' bugger, aren't you
Not to worry, I'm one too but let's keep this civil, OK?

Civil works for me, but seriously, John, you've argued the importance of supersonic information many times. It's a regular theme in your posts. I would think you would have that bit of data, the only bit I recall that has ever supported that argument, at top of mind.

Tim

 

[John Kenny]

No, I didn't read it and doubt you did until I pushed. Otherwise, you would have started off by presenting the bit of information in there that supports your POV. In spite of that support, as a DAC designer, I'd still recommend that you start with the ability to measure and analyze your products. First things first, unless you're just looking for ideas to justify your lack of the proper equipment to do so. After you have your products working, and measuring at absolute state of the art within the audible spectrum, you might start looking at the possibility of improvements beyond.

Tim

OK, so you agree that all your questions have been answered so far.
And now you proceed t make accusations about my lack of equipment based on what?
I thank you for your recommendations & when you design products I'm sure you can put those thoughts into action.
For me, I'll do what I find brings me the most efficient results & if people also find my products worthwhile, I'll use that feedback as confirmation that I'm doing it right