Everything you always wanted to know about sound.... but was afraid to ask

garylkoh

WBF Technical Expert (Speakers & Audio Equipment)
Sep 6, 2010
5,599
225
1,190
Seattle, WA
www.genesisloudspeakers.com
One of my favourite scenes in a movie is this one by Woody Allen - that's why the title of this thread. Just so you know that this is not at all a definitive answer, just the beginnings of a discussion - and I hope that other members will contribute their knowledge.


Our hobby raises a host of challenging philosophical questions - what is it that we are actually hearing? Is hearing spatial? How do our other senses affect what we hear? Both Bruce and I have previously posted this video by Evelyn Glennie where a deaf musician shows us how to listen.

In Hearing Gestures, Seeing Music, researchers Schutz and Lipscomb show that note duration on a marimba is perceived to be longer when the gesture to strike the note is longer even when the acoustic distinction cannot be found. A similar point was made by Ms Glennie in the video where her body language makes the passage she plays sound more expressive.

These studies all demonstrate that ignoring visual information (blind auditions, recordings) rob both the performer and the audience of a significant dimension of musical communication.

However, what then our hifi systems? When we are playing a recording, and there is no visual information of the performance, where do we find that illusion that one performance is more expressive than another? Shouldn't, given what we know now, all marimba/xylophone/vibraphone recordings be flat and boring?

Just listen to John Cocuzzi to hear how much expression and emotion can come into a vibraphone performance even without the visual



Even imaging and soundstage can be influenced by what we see. Both audition and vision are used by the brain to construct spatial maps of the world. But, the eyes are movable and the ears are not, so the brain must accurately account for eye position to maintain alignment. Studies have shown that when the eye position is changed, it can shift sound localization. Auditory Spatial Perception Dynamically Realigns with Changing Eye Position.

In the corollary, with the ventriloquism effect, we perceive sound to come from the location of a visual stimulus. It was found that even monkeys exhibited this plasticity. Just as the auditory signal of an actor's speech appears to come directly from the actor's mouth on a movie screen (even with very badly set up loudspeakers), we perceive the image and soundstage depending on how strong our mental image of the soundstage is.

It has been theorized that music predated language, and that music is evolutionarily important. There have even been studies to show that music can can cause the release of endorphins (and hence relieve pain), lower levels of cortisol (reduce stress and arousal), and raise levels of melatonin (induce sleep). Poets and philosophers have always acknowledged the emotions and sensations that music can elicit.

In a paper to the Conference of the European Society for the Cognitive Sciences of Music, Vincent Meelberg introduced the concept of a sonic stroke - a sound that can create autonomous reactions of the body (like chills up and down the spine). A succession of such sonic strokes bodily invokes meaning and emotions can be treated as causal consequence of bodily changes. Philosophically, in this way, music has an impact on the listener, it induces affects in the body, and we have to explore the ethical aspect of sound.

This understanding serves as a starting point for looking at music where it is not only the listener's mind, but the listener's whole body is the main focus. This leads to Embodied Music Cognition which considers that the human body is a mediator between the mind and the physical environment containing the musical sound.

Finally, what is the nature of "sound" itself? There are three philosophical answers:

1) The proximal theory claim that sound is where the listener is. (If a tree falls in the forest, it does not make a sound if there is no one there to listen.) It is a very egocentric view of what sound is. This also distinguishes between the source and what is being perceived which may be the source, the intervening distance, echoes, reverberations, etc. Hence, there are as many sounds as there are actual (or potential) listeners - lending weight to Ethan's position that moving a microphone as little as 3mm changes the "sound".

2) The medial theory of sound is that it exists between the object making the sound and the listener. (If a tree falls in the forest, it does not make a sound if the forest is in a vacuum.) This goes back to Aristotle, Galileo and Descartes - and is basically the wave theory of sound. Sound exists because of waves in the medium of air - and it is studied in acoustic theory and what we all classically understand. Unfortunately, it is not that simple. Sound waves propagate in all directions from a sounding object (say a musical instrument) and may have a different nature with different direction. Also, a soft sound that is heard close to us is different from a loud sound that is far away. (You could also say that the whole train of soundwaves from the source to our ears is the "sound".)

3) The distal theory considers that the sound is located at the object making the sound. (If a tree falls in the forest, the falling of the tree is the sound.) Hence, sound is a temporal event created by an object. It provides a precise location and time that sound "happens". In this theory, even if a bell is vibrating in a vacuum, it is making a sound - surrounding the bell in air is revealing the sound to the listener. (An apple is still red in the dark, turning on the light reveals it as red.) This gives me the biggest headache - if the sound is a located event at an object (musical instrument), then what we hear in our hifi system is a different sound from a different object - a loudspeaker. When we see an object in the mirror, we are not seeing another immaterial object located in an immaterial space beyond the mirror - there is no such immaterial object, just located incorrectly.

In summary, sounds are here (proximal), there (distal) or everywhere (medial). It has also been denied that sounds have any locations at all - which gives rise to aspatial theories which give me a major headache.

So, dear readers, if I haven't given you a headache yet, in a future installment I'll discuss how some of these philosophies give rise to different ways in which I think as a loudspeaker designer.
 

kach22i

WBF Founding Member
Apr 21, 2010
1,591
210
1,635
Ann Arbor, Michigan
www.kachadoorian.com
It has been theorized that music predated language, and that music is evolutionarily important. There have even been studies to show that music can can cause the release of endorphins (and hence relieve pain), lower levels of cortisol (reduce stress and arousal), and raise levels of melatonin (induce sleep). Poets and philosophers have always acknowledged the emotions and sensations that music can elicit.

I've never read about such studies, but it seems reasonable.

The Woody Allen video was great, I know it had a point but I'm still smiling.

Good topic like Steve said.
 

fas42

Addicted To Best
Jan 8, 2011
3,973
3
0
NSW Australia
Gary, having just reacquainted myself with this posting, and in context with your comments in that other thread, I would put my money on the medial theory, at least from my reading of your descriptions of the various types here. The medial takes into account both the producer and receiver of the sound, there is an "agreement", so to speak, set up between the two entities.

As examples, good sound could be perceived by the listener if he were in an extremely favourable mood, from enjoying fine glasses of wine or a very agreeable companion :))), even from poor reproduction of sound. On the other hand, if the perceiver were in a lousy, kick the cat mood the sound eminating from some device would have to be of a very high order of pleasantness and quality not to cause further aggravation and displeasure ...

Frank
 

garylkoh

WBF Technical Expert (Speakers & Audio Equipment)
Sep 6, 2010
5,599
225
1,190
Seattle, WA
www.genesisloudspeakers.com
The Nature of Sound

This thread is a philosophical “think piece”. It is not a rigorous scientific paper, so please treat it as such. As with all philosophy, there is no right or wrong answer, we can only explore the differences.

Since ancient Greece, philosophers have pondered the question. “If a tree falls in the forest, and there is no one to observe it, does it make a sound?” In order to answer this philosophical question, we have to look at the nature of sound. Are sounds individuals? Are they events? Or are they properties of sounding objects? In the visual corollary, if an apple is in the dark, is it still red?

It turns out that we can neatly classify the various philosophical answers according to the spatial status of the nature of the sound. There are the Proximal theories that state that sounds are where the hearer is. Hence, if a tree falls in the forest and there is no one to hear it, there is no sound. The Medial theories of sound locate the sound in the medium between the object and the hearer. If a tree falls in the forest, and the forest is in a vacuum, there is no sound. This is where a lot of mainstream acoustics consider sound to be – especially in room acoustics.

Distal theories consider sounds to be located at the resonating object. Hence, the falling tree IS the sound, even if there was no one to hear it, and if the tree fell in a vacuum (no medium). Finally, some deny any spatial relevance to sound, and they subscribe to the Aspatial theory of sound.

Sound theories can also be classified according to metaphysical status (are they an occurring event, or object with properties, or dispositions?). There are crossovers and we will see interactions between these different theories of “What is sound?”

Let’s examine each of the four theories, and how it relates to our hobby – audio.
 

garylkoh

WBF Technical Expert (Speakers & Audio Equipment)
Sep 6, 2010
5,599
225
1,190
Seattle, WA
www.genesisloudspeakers.com
Proximal Theories of Sound

The proximal theory of sound construe sound as at the hearer. Sound is directly perceived, and are sensations of some sort produced in the observer when soundwaves strike the ear. The theory that sound is a sensation is further justified by subjective sounds - hearing voices or bells even when no one is speaking and no bell is ringing.

In an anechoic chamber, some people report buzzing, whistling and this has been studied as pathologic tinnitus. Shostakovich is supposed to hear melodies when he turns his head on a side because of a fragment of metal lodged in his brain.
http://www.scribd.com/doc/28033173/Shostakovich-left-side-ventricle-fragment-and-music

However, if sounds are defined as objects of audition, then there must be a different stimulus, and sounds can be defined as the objects of auditory perception. When this position is taken, then the sound is only heard where the hearer is. And the logical conclusion is that there are as many different sounds as there are hearers (or positions where there can be a hearer). It also leaves open the possibility that an unheard sound is located away from the hearer.

This can be further supported by the fact that in the listening room, moving the microphone a couple of mm or tenths of an inch, you get a different frequency response. Hence, the sound is different in different seats. If the musician playing was far away, the physical sensation in the ear is different from if the musician was nearby. The sound is very different.

In this case, the sound contains not only the information on the musician, but also attenuation due to distance (which is different for each frequency due to the restitution of air), echoes, reverberations, filters (other bodies in the audience), etc. And this sound is different in every seat in the concert hall.

However, we can and should argue that there is one musician (or band), one performance, and the sound is the same, just heard in a different way at different locations. The major shortcoming of the proximal theory of sound is that they do not locate sounds where the hearer is perceiving it as coming from - ie. you can see a musician, you can perceive his "sonic image" but sound is actually "here".

If you take the proximal theory of sound, as an audio designer, you will create an anechoic chamber for a listening room, and listen with the loudspeakers directly aimed at the ear because what music you want to hear on the CD should already be on the CD. You don't want all the contributing sounds of the listening room to intrude.

Binaural recordings also are embodiment of the proximal theory of sound. Since sound is at the ears of the hearer, put microphones inside the ear and record what the hearer hears. Then, reproduce these through headphones and you have perfected sound recording.
 

garylkoh

WBF Technical Expert (Speakers & Audio Equipment)
Sep 6, 2010
5,599
225
1,190
Seattle, WA
www.genesisloudspeakers.com
Medial Theories of Sound

Aristotle wrote in the 3rd century B.C. that "sound is a particular movement of air". Scientists in the 17th century (2000 years later!!) refined this intuition into the wave theory of sounds. Descartes consider what we hear as not the objects themselves, but some movements coming from them, and Galileo said "sounds are produced and heard by us when a frequent vibration of air shaken in tiny waves moves a certain cartilage of the tympanum in our ear."

Nevertheless, the Medial theory of sound is not just revisionary - since it must take the wave in the medium to create the sensation at the ear. Nevertheless, it takes the difficult psychological position of auditory perception away from the study of the physical manifestation.

The modern wave theory of sound endorses the medial theory of sound. Sounds are construed as mere mechanical vibrations transmitted by an elastic medium. They are described as longitudinal waves defined by their frequency and amplitude. Our understanding of sound in the medial theory takes on purely physical properties: a vibrating object (musical instrument, vocal chord, etc.) creates a disturbance in the surround air and sets each particle of the air in a back and forth motion at a given frequency at a given amplitude, and this motion propagates to the neighboring particle, but with an energy loss that entails a decrease in amplitude.

Thus, the sound we hear is identical with the train of airwaves that stretches from the distant sounding object to our ear. What we perceive strongly correlates with the properties of waves. The pitch correlates with the frequency of the waves, the intensity (or volume) correlates with the amplitude. The directionality of sound can be explained by the fact that the soundwave train is a propagation line from the source to the hearer.

Unfortunately, in the medial theory of sound with sound waves, there are metameric sounds that feel and sound identical to the ear although the corresponding medial properties are different. For example, you cannot hear the difference between a steady sawtooth wave, and a waveform that contains exactly the same harmonics, but the phase of the harmonics delayed by various amounts. It would not look ANYTHING like a sawtooth on an oscilloscope, but you would not hear the difference.

(As an aside, unfortunately, people use this to argue that you cannot hear phase abnormalities. While the above is true about a steady wave, it is absolutely not true with music. Dynamically changing waveforms with transients are damaged by phase shift - for example changing a "ting" into a "thwock".)

Moreover, ultrasounds and infrasounds have the same physical nature as sounds - but they are not audible, so do they count as sounds? The relationship between the soundwaves and the sounding object is also in question - do sound waves depend on the sounding object? Or are they an event (temporal)? And as with the Proximal theory, the Medial theory also does now well locate sound where auditory perception suggests that the sounds are.

An argument against the medial theory of sound is that the location of the sound plays too large a role in our perception of the sound. A loud sound heard from far away is different from the same sound played softly in closer proximity. The spectra of the two sounds is very different. If you amplify the sound of a singer who is singing quietly, you don't get the impression of the singer singing loudly, but you get the impression of coming close to a quietly singing giant. Thus, it is very important in our set-up process to have "real life levels" when we are setting up and designing our systems. A too-large image could be due to too high amplification.

Another argument against the medial theory of sound is that a bell struck in a vacuum is like an apple in the dark. The apple will still be red when the light is turned on, and the bell will still sound when air is re-introduced.

Nevertheless, the identification of sound with soundwaves is compatible with the fact that soundwaves have an origination, and hence localizable. The way that the soundwaves propagate in the environment can support all spatial theories (proximal, medial, and distal). A further extension of the Medial theory of sound is that there are TWO distinct medium at hand - the source medium (the thing that is vibrating - or sounding object) and the environment medium surrounding the thing that is making the sound and that is propagating the sound. This clearly becomes clear when for example we consider a flute. The "sounding object" is not the flute itself, but the air inside the flute that is excited to vibrate.

How the Medial theory of sound impacts the loudspeaker designer is that the loudspeaker now becomes the sounding object to excite the medium. It translates into a time/location lens that is interjecting between the actual sounding object and the hearer, recreating the wave that reaches the hearer.

Consequently, in order for the reproduction system to recreate the sounding event, it has to capture the soundwave on the way to the hearer, and have these soundwave recreated to continue on its way to the hearer. This seems conceptually a much harder thing to do than binaural recordings recreating the sounding event according to the proximal theory of sound.
 
Last edited:

garylkoh

WBF Technical Expert (Speakers & Audio Equipment)
Sep 6, 2010
5,599
225
1,190
Seattle, WA
www.genesisloudspeakers.com
Distal Theories of Sound

As far as spatial theories go, we've already covered sound being here, and everywhere, we need to cover sound being there. The Distal view identifies the sound as a property, process or event in (or on the surface of) the sounding objects. The Distal view claims superiority over both Proximal and Medial theories by virtue of the fact that there is a spatial structure of the auditory content, and sound can be distally localized.

There are at least four ways to look at the Distal theory of sound, each differing in describing the ontological status of the sound. http://en.wikipedia.org/wiki/Ontology

Sound as a Property
First, sound can be seen as a property of the sounding object - endorsed by Galileo among others. However, there are a number of objections because of the temporal issue. Sounds stop and start, and it is difficult to think that an object having a sound (unlike an object having a color). We say that objects produce a sound, and sound is dynamic. They take up time, start and stop.

Sound as a Located Event
Secondly, sound can be seen as an event that is located at the sounding object, and are identical to, or supervene on the vibration processes of the sound. In this view, the auditory perception of sound still requires a transmitting medium. However, unlike the Medial Theory, the transmitting medium is not essential to the existence of the sound. Sound is intrinsically temporal. They start and cease.

There are a number of compelling reasons supporting the Located Event Theory:

1) Sound does not propagate from the object as much as sounds do not appear to - the sounds can easily to be perceived to be located at the object (imaging in audiophile parlance)
2) The vibration of the sounding object do not move any more than the sounds appear to (when the sound "appears" the sound is a feature of the object
3) Distal volume is constant - unlike soundwaves, the intensity of the vibration can remain the same, even if one one moves away from the sounding object. It is the medium that attenuates the sound.
4) The sounding object continues to vibrate even when the medium is taken away. The medium reveals the sound much like light reveals the color of the apple.
5) Sound is a perceptual deputy of the sounding object. We hear a telephone (and identify it) or a cockroach (and correctly identify it) or a clarinet (and some can correctly identify it).

However, there are a couple of metaphysical objections - if it is a located event, when we record it and transmit it across time and space, is it the same sound? What about an echo? Or reflected sound? There is then the temptation to identify the sound with the soundwave. But if you consider the visual analogy, seeing an object in a mirror id not seeing another immaterial object located in an immaterial space beyond the mirror-plane.

In the Located Event Theory of sound, it is the deviant paths of sound waves as echoes that is responsible for the perceptual difficulty in locating sound. But this is distinct from the view of the Medial Theory that the soundwave IS the sound. The Located Event Theory also supports that "sound fills the room" because the located event generates spreading soundwaves that are then reflected by the walls and the waves seem to fill the room. Unlike smells, when the sound emitting object ceases to "sound" the sound quickly disappears (save for the reverberation trail - which is then the sound of the room.

As an extension, the Relational Event Theory associates the sounding object with the medium sharing with the Medial Theory the indispensability of the medium to the existence of a sound. This argues that we can distinguish between not hearing a sound because there is no medium and not hearing a sound because the sounding object has stopped sounding. The visual analogy is the perception of an object in the dark. When the light turns on, we see the color, and we do not have the impression that where there is no light the object loses its color. Nevertheless, the object reflects light of the color that it is, the medium carries the sound of the object.

Sound as a Disposition
Continuing with the Distal Theory, sound can also seen as a Disposition of a sounding object to vibrate in response to being stimulated. This takes away the temporal dimension by substituting a stimulus. The dispositional theory intuitively helps in explaining that sounds have a loudness (an object is hit harder, or blown harder); assists in explaining transients (we impart a stimulus to the object) so that the sound starts fast, and then fades; and allows an explanation of sound constancy. It also neatly explains that the stimulus is significant for the sound. Just as white light reveals all colors with great fidelity (green grass in the orange light of dusk still looks green, just a weird shade of green), a good stimulus reveals good sound (a good violinist makes a violin 'sing'). However, a violin always sounds like a violin even when the quality of that stimulus isn't great.

There are three dispositions of an object - the disposition to react when thwacked, the disposition to react in the decay mode, and the disposition to vibrate in a certain way. These three dispositions help us when designing a loudspeaker, because they correspond to the impulse response, the waterfall decay, and its resonance. They can also be associated with the way a room responds to sound.
 

garylkoh

WBF Technical Expert (Speakers & Audio Equipment)
Sep 6, 2010
5,599
225
1,190
Seattle, WA
www.genesisloudspeakers.com
Aspatial Theories of Sound

When we assigned sound a spatial location (here, there, everywhere) problems arise and questions about whether sound is really located in space, or it's just the sounding object or the hearer that is located in space. Hence, there is also a non-locational, aspatial theory where there is the plausibility that sound itself is non-spatial. For example, we do not say that sound is "to the left of". We say that the saxophone is to the left of the double bass. We correlate the sound the object, but we can also perceptually represent a non-located sound (the music fills the room).

It is an error theory of auditory perception that we hear sounds coming from where there is no sound source - for example when you hear a sound coming from behind you or to the side even though there is just the two loudspeakers in front of you (like with many Q-sound recordings). Thus, we can perceptually locate a sound without representing its location, and we can perceptually represent a sound that does not have a location. Ventriloquists rely on this dichotomy between the perception of the sound and the perceived spatial localization of the source of the sound.

Sounds as Pure Events

A special category in aspatial theories classify sound as a pure non-physical event and is a secondary object. Thus, sound as music can be heard as an event that is detached from the physical cause of the sound - thus it can be captured and reproduced - but as a secondary object, its nature and properties are determined by how they appear to the person perceiving the object. Thus, music is produced physically but is different and detached from the physical disturbance of the physical objects that produced them.
 

JackD201

WBF Founding Member
Apr 20, 2010
12,308
1,425
1,820
Manila, Philippines
You're one deep dude Gary.
 

garylkoh

WBF Technical Expert (Speakers & Audio Equipment)
Sep 6, 2010
5,599
225
1,190
Seattle, WA
www.genesisloudspeakers.com
Thanks, Jack and Tom. When I started this thread, I was worried that I would lose everybody on the forum. But it also clarifies things in my own mind when I take the effort to distill the concepts down to simple concepts and analogies.

Take for example the "continuity of existence" paradox to cables. If music goes in one end, and music is an object, when it comes out the other end, is it still the same object?
 

flez007

Member Sponsor
Aug 31, 2010
2,915
36
435
Mexico City
Still "digesting" the thread content Gary, this is the sort of reading where one needs to go back again...three times now in my case. Great stuff!
 

JackD201

WBF Founding Member
Apr 20, 2010
12,308
1,425
1,820
Manila, Philippines
Thanks, Jack and Tom. When I started this thread, I was worried that I would lose everybody on the forum. But it also clarifies things in my own mind when I take the effort to distill the concepts down to simple concepts and analogies.

Take for example the "continuity of existence" paradox to cables. If music goes in one end, and music is an object, when it comes out the other end, is it still the same object?

Isn't it that every stage creates a copy? :)
 

garylkoh

WBF Technical Expert (Speakers & Audio Equipment)
Sep 6, 2010
5,599
225
1,190
Seattle, WA
www.genesisloudspeakers.com
Sorry that this thread has taken so long to continue...... Reading the thread mep started about measurements and correlation to what we hear prompted me to continue on this one. What is on the recording vs correlation to what we hear.

Take the argument of live vs recorded. If you were to record a solo guitarist, and then play it back on your hifi system, does it sound the same? How close to "live" is it?

I would ask the question in a different way. How would you record someone playing a guitar and how would you play it back in order for the recording to sound "live"?

If you subscribe to the Proximal theory of sound, you would place a binaural dummy head where you are seated and make the recording, and then listen to the recording using headphones.

If you subscribe to the Distal theory of sound, you would close mike the guitarist, and then place a loudspeaker where the guitarist is, and play the recording through the loudspeaker, and listen where you are.

If you subscribe to the Medial theory of sound, you would place two microphones between you and the guitarist, and play back the recording through two loudspeakers placed where the two microphones were.

Unfortunately, in the vast majority of recorded music, the recording technique and the playback technique are not congruent with respect to the nature of sound.
 

fas42

Addicted To Best
Jan 8, 2011
3,973
3
0
NSW Australia
If you subscribe to the Medial theory of sound, you would place two microphones between you and the guitarist, and play back the recording through two loudspeakers placed where the two microphones were.

Unfortunately, in the vast majority of recorded music, the recording technique and the playback technique are not congruent with respect to the nature of sound.
From that example the Medial looks the go to me ...

Frank
 

garylkoh

WBF Technical Expert (Speakers & Audio Equipment)
Sep 6, 2010
5,599
225
1,190
Seattle, WA
www.genesisloudspeakers.com
Good choice. Aristotle, in his "On the Soul" said that sound was a particular disturbance of the air, and was a supporter of the Medial theory. It then took the better part of the next 1,700 years for scientists to develop the wave theory of sound.

However, there are a number of shortcomings to the Medial theory - the most important to us is that the Medial theory does not allow for imaging and soundstage, which is the Located Event theory, a subset of the Distal theory of sound.

This being a philosophical discussion, there is no right or wrong answer, or even an answer at all.......

Now, back to food :)
 

fas42

Addicted To Best
Jan 8, 2011
3,973
3
0
NSW Australia
However, there are a number of shortcomings to the Medial theory - the most important to us is that the Medial theory does not allow for imaging and soundstage, which is the Located Event theory, a subset of the Distal theory of sound.
Gary, would you care to explain further why you believe imaging, etc aren't part of Medial (after you've had your meal, of course :) )

Frank
 

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. This is THE place where audiophiles and audio companies discuss vintage, contemporary and new audio products, music servers, music streamers, computer audio, digital-to-analog converters, turntables, phono stages, cartridges, reel-to-reel tape machines, speakers, headphones and tube and solid-state amplification. Founded in 2010 What’s Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals, we enjoy learning about new things and meeting new people, and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing