psychoacoustics behind great audio reproduction

[jkeny]

This topic fascinates me & I believe that teasing out the psychoacoustic rules that determine great sound in our 2 channel reproduction chains is the next advance that audio reproduction has to address.

I have not seen it directly addressed on this forum - it is touched on in many spots so I thought it might be interesting to bring together what factors may be responsible for a more realistic 2 channel reproduction.

I'm not interested in fighting over analogue Vs digital or what is euphonic distortion or which is technically superior - I'm interested in gathering together a list of agreed sound improvers & teasing out possible psychoacoustic reasons behind them - it doesn't have to be correct - consider it more a brainstorming session.

This was seeded from Dr. Ulirich "Uli" Brueggemann crosstalk function, called "flow" in Acourate that dallasjustice started a thread on. There are many interesting factors that are being related in that thread which I would like to gather here & maybe add to with other relevant factors on for discussion.

As that thread is mainly about how digital can sound more relaxed by applying a factor found in vinyl playback, I thought it might be interesting to gather together the other factors that vinyl has - some of which may be detrimental to our perception of realism & some advantageous.

This is a complicated area for consideration so let's simplify the discussion - let's consider that the event was recorded live in a real acoustic space & we are reproducing it in our acoustic space using our 2 channel system. So, we immediately have one acoustic space of overlaying another & we are interested in perceiving the underlying one & we have a two point source trying to create an illusion of a 3D space. The most realistic way this can be achieved is to tick as many of the psychoacoustic boxes as such a system allows (let's leave aside any talk of how multi-channel). What I'm trying to gather together here is a possible list of such factors from people who are genuinely interested in improving their listening experience (without the usual arguments)

- Uli's Crosstalk which is frequency dependent seems to give a more relaxed listening experience. Why? Because the theory suggests that it gives us better localisation cues by correcting the slight misalignment that using strictly loudness based localisation causes. This is an interesting anomaly which occurs because we tend to naturally localise both with ILD (loudness differences of the same signal arriving at our ears) & ITD (timing differences of the signal arriving at our ears). When ITD is taken away (not used in the recording process) then ILD becomes less than 100% effective at localisation & leads to some smearing of the location of the sound because just relying on ILD we psychoacoustically don't perceive all frequencies of an instruments signal as coming from the same place. Correcting the psychoacoustic frequency variability of the loudness would seem to be worthwhile to addressing this. One way of slightly correctly shifting the frequency elements of the source signal is to introduce some crosstalk which varies with frequency. This will probably only work for those recordings where studio panning is used & maybe not for the situation of a recording of a real venue done with a good microphone technique (others will have to comment on this as I'm unsure of the recording processes & micing techniques) - perhaps recordings from 50s & older classical recording fall into this category?Maybe most modern recordings are manipulated in the studio in some way?

- it appears that vinyl playback automatically imposes this sort of frequency dependent crosstalk mainly because of the mechanics of cartridges, I believe? Could this be one reason why classical recordings are considered more successful on digital than on vinyl (could be ticking my neck out here but I'm just reiterating what I thought was a consensus - if such a thing can be found in audio).

- If we are considering psychoacoustics it's worthwhile to examine (not necessarily blindly accept) what some experts in the field have to say. I quote here one such J_J & what he has to say about vinyl playback:

But LP can sound like it has more dynamic range, because of the distortion/loudness growth issues. Likewise, it can have a wider, more complex soundstage for the same reasons. Many LP playback systems do enhancement of the L-R part of the stereo signal due to both stylus beam pivot issues as well as cartridge design issues.

So, two issues - apparent psychoacoustic increase in dynamic range & L/R channel enhancement. I don't understand what the last one means - what enhancement, loudness, separation,other?

So this loudness growth issue results in:

More specifically, people hear it as additional positions in the soundstage being introduced at high levels. Not width, which is has also been evaluated, and which of course changes with changes in M/S balance.

I'm wondering if this could be openly discussed without trench warfare?


Some other factors that have been identified on the Uli thread as possible factors that could do with a more widespread airing:
- microphone choice & positioning (close mic'ing being one of the unnatural choices?)
- a small bit of reverb in vinyl playback

Something else that tweaked my interest is the idea of splash speakers like the ceiling firing Zephrin speakers where an attenuated version of the soundfield is bounced off the ceiling so as to represent a late arrival reflection. The cube8 & other multi driver, multi directional firing speakers seem to provide the same sense of envelopment. My first thoughts on this were that perhaps the late arrival gives a better psychoacoustic clue that allows us to continually & easily differentiate the room's playback acoustic from the venue acoustic on the recording?

Anyway, I'm hoping that other ideas & theories/guesses could be offered here as an open discussion area - a brainstorming thread on psychoacoustics & our 2 channel playback systems?

 

[thedudeabides]

Psycho acoustics contains many elements. Time of day you are listening to music, physical state, mental state, as well as the overall quality of your system to reproduce sonic ques that allow one to hear more of the acoustic recording space, including dimensionality, reflections, and the instruments thereof.

It is so personal and so subjective that, IMHO, it is impossible to objectively quantify.

Haven't heard the other speakers you mention but MBL's do very well at providing a sense of "envelopment".

Some like. Some don't. For me, I will not own another speaker brand.

Having said all that, good luck with your thread.

 

[RogerD]

I posted this over on Michael's thread about Uli's digital flow software. I think this page http://www.ambiophonics.org/Tutorials/UnderstandingAmbiophonics_Part1.html explains how psychoacoustics can improve stereo reproduction. In my experience it involves crosstalk cancellation,limiting frequency bands or focusing on certain frequencies,and time domain correction. I use 2 pairs of small speakers in my configuation and my psychoacoustics circuit is passive. The benefits mentioned at the conclusion of my linked article is what I hear. If you can reproduce the same with digital processing with 2 speakers I would think that would revolutionize two channel stereo.

 

[jkeny]

Psycho acoustics contains many elements. Time of day you are listening to music, physical state, mental state, as well as the overall quality of your system to reproduce sonic ques that allow one to hear more of the acoustic recording space, including dimensionality, reflections, and the instruments thereof.

It is so personal and so subjective that, IMHO, it is impossible to objectively quantify.

I'm glad you brought that up, dude because I do think there is a distinction that needs to be made here. I often see this confusion in evidence on this & other forums. What I mean by psychoacoustics is the mechanism that occurs after the signal leaves the ear structure via the auditory nerve. I know this could be suggested as being at the mercy of the vagaries of our state of mind. But I'm more talking about the generally applicable psychoacoustic cues that give rise to how we naturally perceive sound, not the variation in this which could be effected by all sorts of factors - just as what the ears pick up is also dependent on factors which vary it's response. As I said, I'm not interested in these modifying factors, which are exceptions - I'm interested in trying to establish what might be the underlying rules/cues by which normal perception of hearing operates & gives rise to our normal, everyday experience of hearing.

Haven't heard the other speakers you mention but MBL's do very well at providing a sense of "envelopment".

Some like. Some don't. For me, I will not own another speaker brand.

Having said all that, good luck with your thread.

Thanks!

 

[jkeny]

I posted this over on Michael's thread about Uli's digital flow software. I think this page http://www.ambiophonics.org/Tutorials/UnderstandingAmbiophonics_Part1.html explains how psychoacoustics can improve stereo reproduction. In my experience it involves crosstalk cancellation,limiting frequency bands or focusing on certain frequencies,and time domain correction. I use 2 pairs of small speakers in my configuation and my psychoacoustics circuit is passive. The benefits mentioned at the conclusion of my linked article is what I hear. If you can reproduce the same with digital processing with 2 speakers I would think that would revolutionize two channel stereo.

Thanks for the link, Roger, I'll have a read of it all shortly. But one thing that strikes me is that Ambisonics isn't the full answer according to most who have studied it & the fact that it isn't universally acknowledged. It is impressive in the recordings I have heard when listening through headphones. but I haven't heard music playback on it.

Also, I'm hoping this thread could address more than just spatial cues? I'm thinking of the relaxation that is often not associated with digital listening but more with analogue listening. What is responsible for this - something as simple as a drop in energy of HF in analogue Vs digital, or some other factors.

BTW, I'm interested in your psychoacoustics circuit - got a schematic or link?

 

[RogerD]

Thanks for the link, Roger, I'll have a read of it all shortly. But one thing that strikes me is that Ambisonics isn't the full answer according to most who have studied it & the fact that it isn't universally acknowledged. It is impressive in the recordings I have heard when listening through headphones. but I haven't heard music playback on it.

Also, I'm hoping this thread could address more than just spatial cues? I'm thinking of the relaxation that is often not associated with digital listening but more with analogue listening. What is responsible for this - something as simple as a drop in energy of HF in analogue Vs digital, or some other factors.

BTW, I'm interested in your psychoacoustics circuit - got a schematic or link?

Is your name John....thanks. Now this is just my opinion and based on my experience. The only reason Ambiosonics hasn't caught on is people are always skeptical of something new. Frankly I think the majority feel everything is just fine,leave it alone. But the improvement in the stereo chain has improved so much in the last 10 years,maybe now is or near the time that Ambiosonics will have a place in High End audio. You can only squeeze so much out of a limited technology and with digital processing it could have a breakthrough.

Relaxation? That is not the fault of the digital format IME. Ambiosonics will reduce distortion,but that has an affect on analog or digital. There is another reason but that would not benefit this thread.

My circuit is a one of a kind,custom made for me by a very intelligent audio electronics engineer some 30 years ago. I'll see what I can do.

 

[RogerD]

Btw you might find this interesting as others have experimented with stereo. I use the same six speaker configuration except with my time domain circuit it avoids the pitfalls.

http://www.soundfountain.com/mercury/trinaural.html

 

[Kal Rubinson]

- Uli's Crosstalk which is frequency dependent seems to give a more relaxed listening experience. Why? Because the theory suggests that it gives us better localisation cues by correcting the slight misalignment that using strictly loudness based localisation causes. This is an interesting anomaly which occurs because we tend to naturally localise both with ILD (loudness differences of the same signal arriving at our ears) & ITD (timing differences of the signal arriving at our ears). When ITD is taken away (not used in the recording process) then ILD becomes less than 100% effective at localisation & leads to some smearing of the location of the sound because just relying on ILD we psychoacoustically don't perceive all frequencies of an instruments signal as coming from the same place.

It is worth mentioning that ILD is more effective/significant for high frequencies and ITD for low frequencies due to the size of the human head with respect to the wavelengths involved.

 

[jkeny]

Is your name John....thanks.

Yes, John

Now this is just my opinion and based on my experience. The only reason Ambiosonics hasn't caught on is people are always skeptical of something new. Frankly I think the majority feel everything is just fine,leave it alone. But the improvement in the stereo chain has improved so much in the last 10 years,maybe now is or near the time that Ambiosonics will have a place in High End audio. You can only squeeze so much out of a limited technology and with digital processing it could have a breakthrough.

Indeed, I know what you say probably accounts for a large part of ambisonics remaining as a side-issue. I just looked further into the site & see some music examples there which I'll have a listen to. The site has certainly progressed from the last time I looked into it so it's augers well that it hasn't died a death.

Relaxation? That is not the fault of the digital format IME. Ambiosonics will reduce distortion,but that has an affect on analog or digital. There is another reason but that would not benefit this thread.

Sure, it may not be a fault of digital, I was just showing some other aspects, other than spatial location that might also be considered in this thread? Griesinger has done research on concert halls & what auditory cues makes for better intelligibility & envelopment & concluded that "multiple sound sources
(either real or virtual) are necessary for optimal envelopment" but went on to define the rules by which these should operate for maximal effect.

My circuit is a one of a kind,custom made for me by a very intelligent audio electronics engineer some 30 years ago. I'll see what I can do.

No problem, Roger, if it's proprietary, I fully understand.

BTW, I've always liked your sig & believe it captures an often forgotten aspect of audio reproduction. The reduction of noise (of all sorts) has a much larger psychoacoustic effect than a simple cursory glance would suggest.For instance, we are very sensitive to the onset of the signal - the attack transient or whatever it is called. A large part of our categorisation of the sound happens in this first couple of mS & listening to the development of the signal is a confirmation or contradiction of this initial perceptual guess (it is just a guess) - al this is a necessary part of our survival mechanism. Anything which blurs or mask this information can lead to a less relaxed perception - a less natural perception. I think there might be other low level cues that are also needed for most natural perception & it's these I'm also interested in finding out/discussing here

 

[jkeny]

It is worth mentioning that ILD is more effective/significant for high frequencies and ITD for low frequencies due to the size of the human head with respect to the wavelengths involved.

Kal, I believe the current thinking is that at higher frequencies we switch to hearing the ITD of the envelope of the sound rather than the ITD of the carrier frequency?

 

[dallasjustice]

Psychoacoustics Conversation With Audiophile:

Me: How good or bad is the playback speaker/room interaction? That's THE question. Everything else is small stuff.

Audiophile: But, Michael, it's the small stuff that matters for me! I mean it's that last 1% that makes ALL the difference in my system. I mean that power cable changed everything!

Me: Have you ever measured your room's FR and time domain response? Do you own a calibrated microphone?

. . . .

 

[jkeny]

Psychoacoustics Conversation With Audiophile:

Me: How good or bad is the playback speaker/room interaction? That's THE question. Everything else is small stuff.

Audiophile: But, Michael, it's the small stuff that matters for me! I mean it's that last 1% that makes ALL the difference in my system. I mean that power cable changed everything!

Me: Have you ever measured your room's FR and time domain response? Do you own a calibrated microphone?

. . . .

Yes, Dallas, I know that these are considered to be the big areas to focus on but I again believe that too much is made of this.

I know that spaciousness & envelopment are related to reflections & these can be improved & made optimal for a particular room but I also believe that if the room isn't dysfunctional then we can & do separate out the playback room acoustics from the playback itself. If the room is relatively intelligible (& most living rooms are - we can easily hear intelligible conversation in these rooms) then our playback systems will not be wildly effected across the frequency spectrum. Yes we might suffer LF freq humps etc. but our malleable hearing perception can adjust to these fixed frequency artifacts fairly quickly.

In actual fact, I'm of the opinion that we can deal with anomalies that are fixed far more easily than ones that vary so, for instance, noise that varies is a problem whereas noise that is fixed we can listen through. As a result, I'm a bit suspicious of noise shaping (in digital) & the potential that there is for modulating noise.

 

[RogerD]

Psychoacoustics Conversation With Audiophile:

Me: How good or bad is the playback speaker/room interaction? That's THE question. Everything else is small stuff.

Audiophile: But, Michael, it's the small stuff that matters for me! I mean it's that last 1% that makes ALL the difference in my system. I mean that power cable changed everything!

Me: Have you ever measured your room's FR and time domain response? Do you own a calibrated microphone?

. . . .

Yes, Dallas, I know that these are considered to be the big areas to focus on but I again believe that too much is made of this.

I know that spaciousness & envelopment are related to reflections & these can be improved & made optimal for a particular room but I also believe that if the room isn't dysfunctional then we can & do separate out the playback room acoustics from the playback itself. If the room is relatively intelligible (& most living rooms are - we can easily hear intelligible conversation in these rooms) then our playback systems will not be wildly effected across the frequency spectrum. Yes we might suffer LF freq humps etc. but our malleable hearing perception can adjust to these fixed frequency artifacts fairly quickly.

In actual fact, I'm of the opinion that we can deal with anomalies that are fixed far more easily than ones that vary so, for instance, noise that varies is a problem whereas noise that is fixed we can listen through. As a result, I'm a bit suspicious of noise shaping (in digital) & the potential that there is for modulating noise.

FWIW..I have never used any room treatments with my system. Granted since I have 50 inches of subwoofer,I need to tweak that with phase,placement,isolation and so forth. The ambiosonics seem to remove the need for anything else.

 

[dallasjustice]

most living rooms are - we can easily hear intelligible conversation in these rooms) then our playback systems will not be wildly effected across the frequency spectrum.

Hand claps and voices don't work. Everyone NEEDS a mic, IMO. I think most audiophile have very broken rooms. It's a testament to our hearing that we can enjoy music at all in many rooms. Anyone that's gone from a bad listening environment to an acceptable one using conventional methods will likely agree.

 

[jkeny]

Hand claps and voices don't work. Everyone NEEDS a mic, IMO. I think most audiophile have very broken rooms. It's a testament to our hearing that we can enjoy music at all in many rooms. Anyone that's gone from a bad listening environment to an acceptable one using conventional methods will likely agree.

Sure, I'm not ignoring it, just not willing to overplay it either. So, it's not a show stopper for me. In other words I still think that significant improvements can be made to the rest of our systems independently to dealing with room treatments or room correction. I've been in rooms that have been treated & apart from sound studios I haven't noticed a big difference between then & untreated rooms. Maybe there is & I just haven't visited a very well done rooms with professionally designed room treatments. The sound studios I have visited have substantial treatments that wouldn't work aesthetically in a home environment so maybe I'm jut biased?

 

[Kal Rubinson]

Kal, I believe the current thinking is that at higher frequencies we switch to hearing the ITD of the envelope of the sound rather than the ITD of the carrier frequency?

Probably in the crossover region.

 

[esldude]

Hand claps and voices don't work. Everyone NEEDS a mic, IMO. I think most audiophile have very broken rooms. It's a testament to our hearing that we can enjoy music at all in many rooms. Anyone that's gone from a bad listening environment to an acceptable one using conventional methods will likely agree.

I agree. Once you hear a room with some of the primary issues resolved going back also made me wonder how we hear and enjoy as well as we do in most rooms.

 

[jkeny]

Probably in the crossover region.

It's not dependent on speaker crossover, if that's what you mean?
Many references will be found to this statement if you search for "envelope ITD"
"At high frequencies, interaural time differences (ITDs) are conveyed by the sound envelope. Sensitivity to envelope ITDs depends crucially on the envelope shape."

 

[esldude]

Just messing around on a rainy afternoon with the idea about pulse trains or envelopes of high frequencies. Tripped across an interesting phenomenon.

The zipped file has 3 mp3 clips in it. Each is a short 5 seconds. Tones are not deafening in level, but start listening low and set volume to a comfortable level. Two have a momentary 300 hz sine followed by a few seconds of 3khz sine waves. Try them and let me know how they image to you. They image differently to me over headphones. The third is a pair of plucks. I believe you will find the plucks image over to one side.

Now the 300 hz followed by 3 khz tones have one channel delayed by 1 millisecond. But have a minor difference otherwise. Yet image differently to me. Another thing to try with the two 3khz tones is skip over and start the file at the 2 second mark and see if they image differently.

The pluck has the same 1 millisecond delay in one channel and image over to one side even though all frequencies are 3 khz and above.

 

[esldude]

Probably in the crossover region.

Try the short files in my attachment in the above post. I think if you try these with headphones (no crossover) that these files will show it isn't a crossover effect.