Stereophonic Sound -- BINAURAL LISTENING

Home | Audio Magazine | Stereo Review magazine | Good Sound | Troubleshooting

No doubt about it, big words like binaural and stereophonic can be impressive, particularly when used by the man who sells stereophonic sound. Then you get the impression that if you only have ordinary, un-stereophonic sound, you haven't lived yet. So, "All right", you say, "I'll have stereophonic sound", but then you discover that the cost of both equipment and recorded material for stereophonic sound is about twice that of ordinary, un-stereophonic sound.

This makes you think again; "Just what do I get for all this extra cost? In what way does stereophonic sound differ from un stereophonic sound?" The salesman has an answer ready. Most probably he will draw your attention to the difference between ordinary photographs, either black and white or in color, and 3-D or stereoscopic pictures. He will explain to you that ordinary photographs arc two-dimensional, or flat, while stereoscopic pictures provide you with the third dimension, making the pictures appear in "depth", which is what we mean by calling them 3-D. This is a wonderful made-to-order illustration for the stereo phonic salesman, because the difference between two-dimensional and three-dimensional pictures is certainly dramatic. He will tell you that the difference between two-dimensional and three-dimensional sound is equally dramatic. He may then play you a carefully selected and prepared demonstration that he hopes will persuade you into parting with just twice as many dollars as you would for your ordinary un-stereophonic sound system.

In all fairness however, we must admit that while there are plenty of salesmen of this type around, there are also many who will give you a fair demonstration and explanation. The situation remains somewhat unsatisfactory though, because a great many people are out of reach of qualified advisers who could give a lucid explanation of just what stereophonic sound is. And that is precisely why this guide has been written.

That rather obvious illustration of two-dimensional and three dimensional pictures, while dramatic, is not quite a true parallel for the difference between ordinary and stereophonic sound. In the realm of sound there is no such thing as a direct counterpart for the two-dimensional or fiat picture. It is quite impossible to have a two-dimensional sound. I am not saying there is no difference between stereophonic and un-stereophonic sound. There is. But the difference is not to be so easily explained as by drawing an illustration with 2-D and 3-D pictures.

The qualities about sound and listening that give us an impression of depth are quite different from the qualities about light and vision that give us a similar impression of the things we see.

That is why this guide begins with a section on binaural listening: with the object of establishing a clear picture of just how we hear things. Let's just go around in our everyday experience and notice some of the things that our hearing can do for us.

Most people who read this will have two good ears and the re marks that follow are primarily for their benefit. If you happen to be partially deaf in one ear you will probably find that most of the remarks mean something to you, while some of them are a little difficult to verify in your case.

If you are stone deaf in one ear you may still have some sense of direction, but it will be quite different from that described in the rest of this section, and what you read will therefore just be theory--it will have little or no meaning in your experience.

Learning to Hear

Just how good is our sense of direction in hearing? Have you ever tried calling a child's name, or for that matter an animal, and seeing how accurately he turns his head to look directly where you are? At first, when very young, the child or the animal may need to do a little searching, after looking in the general direction, before he finally sees you. But as his hearing becomes better trained his head will turn in the precise direction in which you happen to be.

This lesson is learned of course, with the aid of vision, but when hearing reaches this degree of precision the fact that it is not dependent upon vision any more can be verified by hiding behind suitable obstacles such as a convenient tree or other hiding place. Although he can no longer see you when he turns his head in that direction, you will find that his hearing direction perception is so acute that he still looks in the correct direction. He can probably identify immediately just which obstacle of many you happen to be hiding behind.

This sense of direction is not confined to sounds coming from a horizontal plane or on a level with you. This you will be able to observe by listening to aircraft going overhead, if you are in the middle of an open space, such as a field, or beach, or a large parking lot. You will find that the first notification you usually get of the approach of an aircraft is its sound. If you are interested to see the aircraft, your head then turns to look in the direction from which the sound is coming and you quickly find it with your eyes.

FIG. 1. Because sound takes time to travel, the audible position of a fast moving aircraft is "behind" its actual position.

Sound Takes Time to Travel

If the airplane is one of those slow-flying kites from a near-by airstrip, the position your ears find will be so close to that where you see the airplane that the sound seems to come right from the aircraft itself. If it is a commercial airliner with a cruising speed of 250 to 300 miles per hour, then the general sense of direction will be similar, though you may notice that the sound does not seem to be so closely identified with the aircraft as it did with the slow-flying job.

But if a modem jet comes overhead at a speed much nearer the speed of sound, say 500 or 600 miles per hour, your sense of hearing alone will not help you much in finding the aircraft visually, until you realize that you have to look some distance ahead of where the aircraft sounds to be. The sound of the aircraft will seem to be traveling in a certain direction, and by looking considerably further ahead in that direction you will find the aircraft itself. This is because between the time the sound left the aircraft and reached your ears, the aircraft itself had traveled a consider able distance (Fig. 1) . From this kind of listening you will realize that trained hearing can identify directions of sound to within a few degrees. If you happen to be listening in city streets, however, you may find some difficulty. The aircraft may not be in the direction it seems to be at all, because the sound is reflected from the walls of the buildings that line the street. This is because sound waves can be reflected and a false sense of direction consequently given. If a number of reflections occur at once all sense of direction is lost and the sound becomes confused.

However, sounds do not have to be outdoors for our sense of direction to operate. If someone speaks to us indoors we can just as readily turn to look in the direction their voice comes from, as if they speak to us outdoors. Notice too that you can very easily locate a clock by the sound of its ticking.

But not all sounds can be so readily located. For example, have you ever been in a gas lit auditorium where one of the gas jets whistled, and tried to locate which jet the whistle came from? You can walk around the building and get no positive identification of direction at all. Eventually the simplest way seems to be to turn off the gas jets one by one until the whistle disappears.

A more modem illustration of this point might be a building heated by steam radiators, where one of the radiators emits a whistle. From the sound you might think that all the radiators whistled, but by turning the radiators off one at a time you will probably find that the whistle emanates from just one of them.

Your sense of hearing however, gives you little help in finding which one.

How is it that the source of direction of some sounds can be so readily identified, while others seem to be so confusing? The reason for this sheds considerable light on what we need to get effective stereophonic sound, so we shall discuss it at greater length later on in the guide. Meanwhile let's pursue a little further the things our binaural, or two-eared, listening enables us to do.

Selective Hearing

So far we have concentrated on obtaining a sense of direction from the sounds we hear. But our hearing faculties can differentiate on other bases than just direction. Have you ever talked to some one in a crowded room, where many other conversations were taking place at the same time? Next time you're in this position, try relaxing your attention for a moment.

You will realize that the background of conversation going on is such as to be completely confusing, unless you concentrate on listening to the person with whom you are talking. Your hearing faculty enables you to single out your friend's voice from all the other voices in the room. If you listen carefully you will realize that this is achieved by three separate factors acting together: (a) Your sense of direction in hearing is concentrated toward your friend's voice; (b) Your eyes may be helping you to some extent by a degree of lip reading; and finally--(c) Your hearing faculty provides a kind of selective quality whereby your friend's voice is separated out, by his own personal individuality, from all the other voices.

For another example of the power of the hearing faculty to discriminate between sounds, consider listening to an orchestral concert, either live, or reproduced over a really high quality re producer. You can quite readily identify sound from different instruments in the orchestra. They stand out clearly, each with its own individual characteristic sound.

Even at a live concert, from where you are sitting in an auditorium, you may have difficulty in locating the position of the violins by the sense of sound alone because of reverberation in the building. But you will have no difficulty' in distinguishing the sound produced by the violins from sounds produced by other instruments, such as wood-wind or brass.

Fig. 2. The frequency range used by different musical instruments overlaps almost completely, so we cannot separate the sound from individual instruments by distinguishing their frequencies.

Although it utilizes the same range of component vibration frequencies that other instruments use, each kind of instrument has its own characteristic sound. An analysis of the frequencies used by different instruments (Fig. 2) shows that it would be impossible to isolate sounds made by one instrument, or kind of instrument, from all other sounds, merely by selecting certain frequencies. Somehow the hearing faculty must identify the composite sound from whichever instrument attention is given to.

Another example of extreme selective listening is the auto mechanic who can listen to the composite of weird mechanical sounds that come from an automobile engine, isolate any particular group coming from the valve mechanism, the crankshaft, or whatever part he is giving his attention to, and locate the fault quite readily. To the untrained ear the sound may be almost identical with that from any other motor of the same type; the little sound the mechanic hears may seem inaudible.


These are all observations from everyday experience that any reader can verify for himself. But there is another kind of separation that takes place. We have all subconsciously experienced it, but it is a little difficult to demonstrate directly. This is the separation of original sound from reverberation, which is another name for sound produced by reflection or echo effects.

In a large, reverberant auditorium, the echo is quite noticeable, separate from the sound that causes it. Outdoors, where there may be an echo rock effect, this is also true. But in the average living room, or in smaller auditoriums where the acoustics are reasonably good, the reverberation is not noticeable consciously as a separate entity. Our hearing faculty does however, subconsciously separate the original sound from the reverberation.

This can be indirectly but effectively demonstrated with the aid of a good quality tape-recorder, a microphone with good frequency characteristics and some musical friends. Get the friends to per form on one side of your living room (Fig. 3) , place a microphone approximately in the middle and listen on the opposite side while they do so. You will have an impression of the musical program without any noticeable reverberation. The room is much too small for that.

But now make a recording and play it back. The amount of reverberation you will find on the recording has to be heard to be believed. It sounds as if someone had not only removed all the furniture, but also made the room considerably larger, so as to get definitely more echo.

FIG. 3. This experiment (see text) demonstrates that our binaural hearing faculty subconsciously "ignores" reverberation in normal listening conditions, something no reproducing system, even stereophonic, can do.

FIG. 4. A portable recorder with dummy head to hold two microphones, for making home binaural recordings.

The reason for this difference is that the one microphone has to receive all the sounds, original and reverberant, and it is reproduced over one loudspeaker. The original sound you heard by means of two ears, which enable you to hear it subconsciously distinct from its reverberation. The hearing faculty then, because this is a characteristic of the room that is always with you, sub consciously reduces the noticeability of the reverberation.

The microphone has no means of doing this, and once the pro gram has been recorded and reproduced your hearing no longer has the means to do it either, because the sound, which was previously interpreted separately by each ear, has now become all mixed up.

Much of the skill with which our hearing faculty can discriminate between different sounds, arises from the use of two ears to pick up separate samples of the complete sound pattern and pass them on to the brain for a sort of computer analysis, which we subconsciously perform all the time.

Referring back, if instead of just one tape-recorder you get a twin-track recorder, and two microphones spaced apart about the distance between an average pair of ears--with some kind of obstacle between the microphones to represent an average head--and then make a recording, you will be able to reproduce this so it sounds like the original program. One way of getting the right pickup is to use a dummy head, with microphones mounted in side (Fig. 4) . Reproduction is achieved by playing back each recording into one of a pair of headphones so that each ear of the listener hears what was received by the microphone corresponding to that ear in the original performance. Although each channel in the tape-recording has just as much reverberation recorded along with it as did the original single-channel recording, we have now provided the hearing faculty with the means for achieving satisfactory separation. Listening to both at once, the reverberation no longer sounds excessive but seems about natural, as it did when listening to the original sound.

Binaural Reproduction

Here, then, we have the original basis for the whole idea of binaural listening and stereophonic sound. Binaural reproduction, as it is called, uses a pair of headphones and a dummy head, as just described. But the method has two principal shortcomings.

Although the program reproduced over a good pair of headphones, gives a much better sense of realism in this particular respect, it lacks realism in two others.

Firstly, when the sound originates from a pair of headphones the ear does not behave in the same way, or give the same impression to the brain; one rather has the impression of the sound being "piped" to each ear instead of the ears being free to pick up sounds out of space. In addition to this effect on perception, there is the discomfort of wearing headphones throughout an entire performance.

The second major shortcoming arises from the fact that the recording is made with the head of the dummy in a fixed position--the microphones and the dummy head to which they are attached do not move the way the listener is free to move his head--whereas if you move your head at all while listening to the program the effect you get is that of the entire auditorium with its program rotating along with your head (Fig. 5). This is quite unnatural, and the only way to avoid it seems to be to wear a kind of head clamp or have the headphones clamped to the back of the seat in which the listener sits.

Fig. 5. One disadvantage of binaural reproduction with headphones so that if you move your head the whole apparent program source, realistic a, It may be, appears to turn with your head.

Because of these drawbacks of binaural reproduction using headphones, it seems necessary, to get a practical sense of realism, to use different loudspeakers and generate a complete wave pattern in the room. If the loudspeakers could be so placed that each ear of the listener heard only the sound coming from the loudspeaker intended for that ear, the problem would be solved. Unfortunately there is no such simple solution.

Once the sound is released into the room by the loudspeaker, the sound from that loudspeaker travels all around the room. So the sound intended for each ear travels all around the room too, and both ears hear something of the sound intended for the other ear, as well that recorded specifically for itself. This problem and how it is overcome however, we will discuss later in the guide.

For the time being, before we leave the subject of binaural listening, it is important to understand a little better how the hearing mechanism operates to give us the impressions we have noted so far in this section.

How Human Hearing Works

Audible vibrations, which we call sound, are fluctuations in air pressure at a rate or frequency which determines the pitch or tone of the sound heard. The lowest audible frequency is in the region of 20 cycles or vibrations per second, while the highest is somewhere around 16,000 cycles. These figures are average for a number of people; yours may vary from these figures. The lower limit may be between 18 and 22 cycles, while the upper limit varies, with individuals, from 8000 to 20,000 cycles.

FIG. 6. Construction of the outer and middle ear.

The fluctuation in air pressure for sound of average intensity is not more than about 1 millionth of the actual pressure of the air. This means the ear has to be extremely sensitive to minute pressure variations within this frequency range.

First of all, the sound vibrations, all of them, pass along the outer tube from the ear to the ear drum (Fig. 6). This is an extremely thin, stretched membrane which vibrates along with the air in contact with it. On the inside of the ear drum is a small air cavity, the pressure of which is equalized with the outside air by a very small tube called the Eustachian tube, that connects to the outer air again through the mouth. The narrowness of the tube and the fact that it communicates with the mouth, prevents any of the sound energy from bypassing the ear drum, and ensures that the air vibrations we call sound waves cause the ear drum to vibrate in accordance.

This vibration is next communicated to the liquid inside the cochlea (inner ear) by a sequence of three very tiny bones, or ossicles, called the hammer, anvil and stirrup, inside the cavity called the middle ear. The purpose of these bones is to convey all the vibrations of the ear drum, which come from air waves, to vibrations of the oval window, which sets up waves in the liquid contained in the cochlea.

This liquid is contained in two long channels "wound" into the form of a spiral, and separated by a thin membrane known as the basilar membrane (Fig. 7) . The whole make-up of the cochlea is contained in a hard, bony, snail-like structure, so that bodily movement of liquid at the oval window, caused by vibrations transmitted from the stirrup bone, must eventually produce corresponding movements at the round window, in the opposite direction.

The liquid may move all the way up the basilar membrane and back down the other side, or it may move only a short way up one side of the membrane before pushing the membrane aside, thus passing the motion to the liquid on the other side.

At just what point it causes the basilar membrane to vibrate depends on the frequency of vibration: the higher the frequency, the shorter distance the liquid moves before it causes the membrane to vibrate; the lower the frequency, the further up the membrane will be the point of vibration.

The place where the membrane vibrates causes stimulation of the nerve cells associated with it, and this in turn is carried to the brain by one of the nerve fibers in the auditory nerve. The auditory nerve has between 20,000 and 29,000 ganglion nerve fibers, each of which carries "information" about one particular frequency of sound being heard.

FIG. 7 Cross-section through part of the cochlea, or inner ear.

Careful research into the mode of transmission along the nerves between the ear and the brain shows that this information is passed along in exactly the same way as the nerves in the rest of the human body work, by a system of pulses.

Human nerves do not transmit continuous electric currents, but a succession of pulses, each of uniform size. The brain interprets all information received along the sensory nerves by analyzing how many pulses are received and noting along what fibers they arrive--a combination pattern, or immeasurably superior Morse code or teletype system.

A loud sound does not produce a stronger impulse along the appropriate nerves than a soft sound; it produces a greater number of pulses. If the vibration is too feeble to stimulate even one nerve pulse, then it is not audible. This determines what is called the "threshold of audibility". When a sound has become audible it is able to produce one pulse along the nerve about every fortieth of a second--to give the impression of the quietest possible continuous sound.

Further increase in its intensity or loudness will eventually cause a second pulse to be transmitted in an interval less than one fortieth of a second. The rapidity with which the pulses are transmitted along the nerve fiber representing a particular frequency, indicates to the brain how loud that frequency is (Fig. 8). The auditory nerve terminates in a section of the brain similar to that where all other sensory nerves terminate. The nerves tell what is going on around us. In the same way that the pattern of nerve impulses received from our fingers can tell us the shape and texture of an object we may be feeling, the pattern of pulses received along the auditory nerve can tell us all we want to know about sound we may be hearing.

The grouping of individual nerve fibers along which impulses arrive at the same instant, together with the way in which the impulses speed up or slow down to indicate the change in intensity of different components of the tone, tell us what kind of instrument or sound we may be listening to. One grouping indicates a vibrating string; another indicates a wind instrument, where the tone comes from air particles vibrating in the mouth of a horn or an organ pipe.

The way the individual component tones vary, causing difference in the pulse rate, tells us whether the string has been plucked, bowed or struck. We could go on describing the different qualities about sound that can be identified by listening, but the possible variety in the way impulses can arrive over some 20,000 to 29,000 nerve fibers, is virtually infinite.

One point however, is essential here in helping to decide how important are different properties of reproduced sound. Because so many nerve fibers are used to cover the audio frequency range, human hearing is very sensitive to difference in pitch, or frequency.

A small fraction of a musical semitone change is easily detected.

On the other hand, intensity or loudness registers by the nerve pulse rate along the fibers. This means that only by critical listening can a change of intensity of 10% or even 20% be heard. Our experience has come to associate certain groupings with which we are familiar with the particular sounds we recognize.

This is really part of the action of our memory. We remember these sounds by patterns they produce and recognize them when we hear them again. This is the conditioning of our hearing faculty.

Fig. 8. Pulse rate along an Individual fiber of the auditory nerve conveys the intensity of the sound; the identity of the fiber tells the frequency.

Realistic Sound

Without having heard similar sounds previously, a new kind of sound means nothing whatever to us and sounds completely strange.

Sometimes strange sounds may bother us, while at other times we may ignore them. This depends on whether we happen to be interested in something else at the same time.

For example, in the early days of radio and phonograph reproduction, distortion was enormous by modern standards. But at that stage we were only interested in hearing the program, and consequently all the strange sounds caused by distortion went unnoticed. Our attention was focused on the program material and our hearing faculty successfully ignored everything else. We even commented that the reproduction was life-like.

The hearing faculty however, has to contend with every sound the ear picks up--at a high enough level to be audible--whether we are conscious of it or not. So these strange sounds caused by distortion exercised the ear, though we were not aware of it. AU we felt was a sense of relief when we removed the headphones and just talked to friends in the room. Later on we came to the conclusion that the reproduction was not as realistic as we thought it was.

Further attention and familiarity helped us to recognize the distortion components that should not have been there, and the fact that they marred the clarity of the reproduction. This general awakening is illustrated by the fact that some years ago it was "proved" at a demonstration that no one could hear less than 5 % distortion. More recent tests have found that more than two-thirds of the people present could hear 5% , and one-third could detect 1% . A little thought about the things we have discussed in this section will show how extraordinarily flexible the interpretive faculty of the brain really is. The effort of listening to our friend in a crowded room where many conversations are going on at once, makes our brain subconsciously turn down the volume on the general chatter.

The motor mechanic checking a fault in the engine, unconsciously turns down the volume on everything except the particular sound he is listening for.

When you pick up the sound of that aircraft overhead you may be near a highway on which there is considerable motor traffic noise that you can't fail to hear. The frequency components of the motor traffic undoubtedly overlap the frequencies radiated by the airplane. Yet somehow your hearing faculty ignores the traffic noise for a moment and concentrates on the particular sound of the airplane to determine its direction.

In any orchestral program different sections of the orchestra play different musical scores to make up the complete orchestration as the composer wished it to be heard. Listening, and as you be come more interested in the music, you find yourself concentrating on the score as played by one or other section of the orchestra--the first and second strings, the wind instruments, even the drums.

You will find you can hear quite distinctly the separate parts that each plays, although they will be made up of frequencies that overlap one another. This is because your hearing acuity has become use to grouping together the impulses coming from each particular instrument or group of instruments. So you can concentrate your attention on the first or second violins, for example.

Being able to do this however, depends upon the clarity with which the sounds are heard. If the musical components from the different instruments become all mixed up and distorted, so that the tones coming from each cannot be separated properly by the hearing faculty in the brain, then some of our appreciation for individual parts of the score is going to be marred. We may not even be able to separate the parts played by the different instruments as we should.

This is the reason for present day stress on reducing distortion in high fidelity equipment. Having brought it down to the lowest possible level however, something is left wanting and this seems to be a sense of realism, or presence, due to the ability of our two ears to distinguish between original sound and reverberation, and between sounds coming from different directions. When all the recorded sound goes onto one channel and is reproduced from a single loudspeaker, our hearing faculty is deprived of some of the distinguishing means present in live sounds or original sound.

Not that the sound we listen to is two-dimensional. The sounds from the loudspeaker still come out and fill the room, but we do not get the same kind of sound as would have come from the original orchestra or whatever program we are listening to. This is what led to the introduction of the various forms of stereophonic sound. But before we get to that, we need to understand some other things. In this section we have discussed the faculty of hearing and the part binaural listening plays in contributing to it. In the next section we shall straighten out a few things about the nature of sound itself and then we can see just what stereophonic sound really is.

(Adapted from: Stereophonic Sound (1957) by Norman H. Crowhurst)

= = = =

Prev. | Next

Top of Page    Home

Updated: Sunday, 2020-04-19 9:45 PST