A listening technique and its application as a
                    useful method of describing tonal quality

Great tone is like hitting 60 home runs a year for life. Great tone one is picking and choosing what job you want. Great tone is making lots of money. Got world-class tone? You have it all. World-class tone, Tone with a capital “T” opens all doors.
What are the three rules of tone?

1.) You ought to have Tone. 2.) You should have Tone. 3.) You Must Have Tone!

Don’t have world-class tone? Do you want it? Of course you do. How do you get it if you can’t afford a world-class instrument? Read on.

This paper introduces a method for discussing the tone of any musical instrument among those who listen to, play, repair, sell, or manufacture that instrument or accessories for it. What makes a great tone? How is one tone different from another? Exactly—minutely—how? You have to be able to talk about and understand tone on a sophisticated level to be able to map out the most direct route to “there,” wherever “there” is to you.

Christopher Columbus had just a vague idea of where to go. West. It wasn’t a very sophisticated concept was it? Unfortunately, it seems that this kind of sophistication can be found in the world of the musician too. I was discussing the apparent lack of sophistication in the curriculum of musicians concerning tone at a convention recently. A teacher at the table said, “Oh we teach tone. We teach them not to squawk.” This is an example of a binary solution set: (squawk, and not-squawk). Hummmm. This leaves a little bit of leeway in the not-squawk category. Does this mean all those who don’t squawk have world-class tone? I wouldn’t think so. How about you? This is of course, an extreme example, but it can’t be the only one of its kind.

There has been, and is now, much research being done on the concept of tone, or as some express it, timbre. However, most researchers in the field of acoustics, psychoacoustics, psychology, etc. are looking to an overall theory of the perception of timbre in the context of any and all instruments. The focus of this paper is to discuss a working methodology and listening technique for musicians, designers, manufacturers, and others to communicate thoughts and observations concerning tone/timbre in the context of whatever musical instrument type they wish—one instrument at a time (for instance, the violin).

I’ve found that some use the word tone and some use the term timbre when talking about the quality of a musical sound. Differentiation between these two terms, tone and timbre, if one only took into account the usage in scientific papers, would be an easy matter. It would seem that the term timbre is preferred in these papers when the discussion pertains to quality, and that this progression is in force: tone has timbre; timbre has quality [2]. However, tone and timbre have been observed to be used interchangeably by most professionals and amateurs. Both terms are considered to inherently ‘contain’ quality, and because of this, they will be used interchangeably in this paper also.

There are many examples of wonderfully rich prose that reveal how music and tone enhance our lives through their connections to our emotions. However, if feelings were all that one had to categorize the things we find in this world, our lives would be very different. Imagine, and pardon if you will, the description of the parts of a car if one were only to talk in terms of emotion. Really wide and long & thin parts with lazy curves, and lumpy little parts (the hood and the knobs on the radio)—long skinny textured little metal cylinders, different but the same (the different bolts used to hold things together)—rounded lattices (the dashboard’s outer skin)—so on and so forth. How would you go about ordering one? (I’d like a really big long lazy thin one in a pleasing magenta.) What differentiates the one from the other? (No, it needs to be less lazy, a bit longer, and that’s an angry magenta—not pleasing at all.) Objectivity would give you some hints: what year, what model, what color code, how many threads per inch, what length—all these things that are the outward trappings of objectivity make it possible to talk about material things without ambiguity. Shouldn’t it be possible to do the same when we want to talk about tone?

Tonespeak is a flexible method for objectively discussing the tone of any musical instrument. It works. It is the result of years of discussing tone on as a sophisticated level as possible with working orchestral musicians in the field (string, brass, and woodwind). It was commonly found that the usual subjective terminology used in describing tone quality did more to confound the transference of knowledge between designer and musician than it aided. A method gradually arose that based the musician’s subjective terminology of tonal quality with objective observations on where in an instrument’s tone there was more audible energy, and what this meant to the musician and the audience. The result is a useful technique for describing tone/timbre that can be used with any instrument, from any family, amplified or not.

The key is of course, what do people think about a particular timbre? Not all think alike. No surprise there. The mud in the water has always been terminology. Does my mental definition of “muddy” match yours? The problem has been that often it does not. I often discuss terminology with musicians, and one day whilst doing so, I was lamenting the fact that one musician could call a tone “blue” for instance, and another not have the slightest clue what the other meant, because “blue” meant something entirely different to them. Many moons later, my musician friend commented to me that he had immediately thought to himself after I had spoken my lament, that I was making a mountain out of a molehill. That is until one day that when he and another musician in a small ensemble had been unable to come to terms with a particular desired tonality because they just couldn’t communicate what their particular tonal vocabulary meant to the other person. Their particular vocabularies were too personal, too subjective.

A central ideal for Tonespeak is to remove this subjective veil from the vocabulary of tone. Tonespeak is in effect an objectively based, pattern recognition system: “A complete pattern recognition system consists of a sensor that gathers the observations to be classified or described; a feature extraction mechanism that computes numeric or symbolic information from the observations; and a classification or description scheme that does the actual job of classifying or describing observations, relying on the extracted features.”[1]
In this case the sensor is the human ear/brain combination; the feature extraction mechanism is the Tonespeak listening technique described here in Part 1; the classification scheme is comprised of the definitions of the various words used to describe tone.

The tone or timbre of a musical instrument is the result of its producing sound; the one follows from the other. The American Standards Association defines timbre as "that attribute of sensation in terms of which a listener can judge that two sounds having the same loudness and pitch are dissimilar". This definition, in various versions has been in use for over 90 years, and was first put to page by Rieman in the Dictionnaire de Musique in 1913 [3]. However, a musical instrument’s tone/timbre is also a qualitative element of its sound.

The quality of a tone is judged and graded through use of a mental framework of ideas about quality. This framework is subjective (based on your personal views) and can vary widely from person to person.

A mental framework encloses a number of ideas, and its “volume” can be called a domain (a ‘territory, ‘or field of study). This mental framework is also called a ‘perceptual construct,’ or just ‘construct;’ quality is then a construct whose verbalizations (the words used when talking about quality) are the result of making an analogy of one concept in this case a multidimensional [4, 5, 6, 7] perception (timbre) in terms of another concept.

This is called ‘metaphoric cross-domain mapping,’ and it is a principle we humans inherently use as a mechanism for understanding [8]. For example, both historically and automatically this principle has been used as a vehicle for understanding and expressing the domain of quality, where we metaphorically associate it in terms of another domain such as the domain of journeys (getting-somewhere-you-want-to-be=better, not-getting-there=not-better, going-in-circles=no-positive-change=not-better, etc.).

From even this brief example of metaphoric tone descriptions (such as: getting there = better), it should be recognizable that the subjective perceptual construct (the mental framework, domain) of ‘quality’ in regards to tone/timbre has always been talked about in ways that attempt to conquer the slippery slope of subjective terminology with the intent to understand, describe, and define this multidimensionality we know as tone. (It was a figurative and literal statement, needing a full chest of air to navigate.)

In a similar vein another has commented on the attempt to describe tone: “…listeners are asked to produce unambiguous responses on matters for which language provides an extraordinarily rich set of blurred definitions.” [9]

In addition, while our mental frameworks (perceptual constructs) concerning quality can be simple in nature or complex, they usually can be stated in terms of one or two words. In this case they guide a listener to judge whether or not a timbre is, for example, good. Of course, what may be good to one listener may be poor to another. While this type of perceptual differentiation may be thought to be more common in the case where one of the listeners is a student and the other a professional, it frequently occurs even at the professional level where there are conflicting schools of tonal preference. For example, there are several schools of thought in the professional community concerning the tone of cellos, where one school professes that ‘darker’ is better. (It then behooves one to know which school your audition committee is committed to.)

A huge amount of research has gone into automatic systems for the purpose of categorizing recorded musical sounds, especially in the computer music field. A major finding of this research is that spectral centroid methods for correlating the textural descriptor brightness are demonstrably successful [10, 11]. (‘Spectral centroid methods’ utilize statistics and computers after the data is compiled (recorded sound/music is processed according to some criteria). A ‘textural descriptor’ is a word that describes a particular tonal color or quality such as the word bright when someone says about a particular tone: “that’s very bright.”)

However, this research isn’t of much practical use to the working musician where one often needs to be able to tell the difference between good brightness and bad brightness, and in the moment, i.e. right now. This differentiation is often difficult to speak about in a clear manner, even amongst musicians who have no trouble telling the difference between good and bad. This is an example of the gulf between tacit and explicit knowledge. This gulf between a musician’s internalized knowledge of tone and the lack of codified, objectively based verbal descriptions of tone, which are at the same time assessable to the working musician, is the reason for Tonespeak’s conception.

It is unfortunate, but far too many professionals put tone in a priority basket far behind technique, timing, and pitch accuracy. This situation is naturally continued by the teacher who automatically passes along to the student, his or her priority bias concerning tone. In this way far too many students and professionals fail to develop vocabulary sophistication pertaining to tone much, if any, beyond the use of simple labels which are often metaphorically based such as fat, thin, better/not-better, or squawking/not-squawking.

In actuality, tone/timbre is a factor in all three of the aforesaid categories: technique, timing, and pitch accuracy. Tone and an instrument’s playability are directly inter-related. This is something that my tone researches have proven time and again; an instrument’s ease in creating a note, its ability to speak, is directly affected by, and a function of, that instrument’s tone. If an instrument doesn’t speak with sufficient ease, a musician will have to fight with his or her instrument, whether in making the note to begin with, or in hearing it afterwards. This will then interfere with all aspects of a musician’s performance, which must, of course include timing, technique, and pitch accuracy.

Tonespeak is a natural alternative grouping technique to metaphoric grouping/mapping. It is a natural grouping technique because we as a matter of course group things together as part of the constant process of dealing with the huge amount of sensory input we deal with on a daily basis. With the Tonespeak method and techniques, a much improved understanding of tonal quality can be realized, and because of this, communicated in a much easier to understand form than ever before. By interrupting the automatic metaphoric mapping of the perception of an instrument’s timbre onto a subjective concept (i.e. such as making the analogy getting-there = better), and by requiring an objective intermediate step, timbre becomes a much easier subject for speaking clearly about/discussing with others.

The first step in the Tonespeak method entails a deconstructive listening technique where the initial assessment of the tone in question is objective rather than subjective. This first step has to do with the tone of a single note played by any instrument. It breaks up that note’s tonal spectrum into a number of partitions (‘de-constructs’ it), for instance into two partitions - highs and lows, each partition having its own particular strength/loudness. However, two partitions are too few for a meaningful degree of sophistication for the definition, articulation, and communication of even the major categories of tonal textures found in the vast array of musical instruments available today.

Imagine if you will a graphic equalizer with only the overall bandwidth necessary for a particular instrument. “Bandwidth” refers to the range of frequencies that a musical instrument plays. The bandwidth of a violin is obviously different from the bandwidth of a double bass.

Each instrument is to be treated individually as far as bandwidth, but all of these mental instrument-specific ‘graphic equalizers’ will have the same number of frequency bands/segments, or partitions. Say we break this range of frequencies into three (3) groups (partitions): highs, lows, and mids. The highs of a tuba will obviously be different than the highs of a piccolo. Similarly, the lows of a violin will be different than the lows of a trombone. But, every instrument will have highs, mids, and lows.

Our job is then to listen to an instruments tone, and mentally figure out which frequency partition(s) have more energy/loudness than the others, i.e. which of the three knobs on our ‘graphic equalizer’ are higher than the others. For instance, are there more highs than lows? Are there more mids than lows? Are there more lows than mids or highs? All these are the kinds of questions we would want to answer.

Once this initial assessment is made in terms of objective quantities, then the selection of which word or phrase (textural descriptor) best describes this particular tonal texture will take place on much more solid ground. The number of partitions necessary for a useful degree of sophistication without becoming overly cumbersome will be the aim of the next several paragraphs.

Every instrument has its own range of notes. Each of these notes has a fundamental (f) and multiple overtones (also called harmonics) which are mostly integer multiples of the fundamental (2f, 3f, 4f, etc.). The fundamental, also called the first harmonic, is often the loudest part of the sound of a note, but not always. While some researchers include attack and decay as significant participants in timbre, others feel the steady state parts of an instruments sound are sufficient [12]. The loudness of each individual overtone/harmonic is in any event a major characteristic of tone. The tone of any note is developed as the result of the strength or loudness of the fundamental added together with the strengths of the various overtones. (The fact that we don’t hear a bunch off individual harmonics but do hear a single “tone” when an instrument plays a note is the result of a mental process called fusion.)

This of course means that each note has a range of frequencies that are produced whenever that note is played. The fundamental of a note is at the lowest frequency of the tone; the upper harmonics have the highest frequencies; in between the two are the middle harmonics and frequencies associated with that tone. The lowest note that an instrument can play has a low part, a middle part, and a high part to its tone, and so in the same way the highest note that that instrument can play also has a low part, a middle part, and a high part to its tone. Thus every note has a low part, a middle part, and a high part, i.e. these three partitions we’ve been discussing for some time now.

If we were then to talk about a tone in terms of the relative amount of energy in each of these three partitions of the tonal spectrum, we would have a descriptive methodology that would give an objective basis to such common labels as fat and thin. The term fat could be used to describe a situation where the low part of a tone was at least slightly greater in energy level in relation to the other two parts of that tone. However, with the addition of just one additional tone word such as muddy, which also concerns an augmented low-part energy level, muddy would then either have to be synonymous with fat (it isn’t) or else this would constitute a failure of this three partition method of tonal description (it is).

By breaking these three partitions of the tonal spectrum (highs, mids, and lows) themselves into thirds, (three groups of three) we will now have a total of nine areas: lower-lows, middle-lows, upper-lows; lower-mids, middle-mids, upper-mids; lower-highs, middle-highs, and upper-highs. These partitions will be effective in describing the tone of the smallest violin to the largest double bass, because we are only considering the lows, mids, and highs of a single note of a single instrument at a time.

At this level of partitioning of the tonal spectrum, for instance, and for instruments producing real bass notes, discrimination between the terms fat and muddy becomes possible. If we define fat as a distribution of bass energy which favors (has increased amounts/loudness/spl) the middle and upper-bass in relation to the lower-bass, and similarly if we define muddy as a distribution that favors the lower-bass in relation to the mid and upper-bass, we now have a distinction that is no longer subjective. Thus this nine-partition Tonespeak listening technique can be utilized in defining quite sophisticated subjective human perceptive constructs. However, those wanting to utilize this type of descriptive mechanism must develop the skill to tell one partition from another, by listening.

The skill level necessary is gained through a combination of practice and an open mind. The “practice” part should be an understood quantity, whereas “an open mind” may not be.

In consideration of the importance of an open mind, we will first look at the central approach of Gestalt psychology which has to do with simplicity and its effect on pattern recognition: people intuitively prefer the simplest, most stable of informational groupings...what’s simplest--to them, on a personal and judgemental level. “…perceptual entities, such as the solid volumes and empty spaces we perceive around us, are easily confused with real objects and spaces in the objective external world. The illusion of perception is so compelling that we mistake the percept of the world for the real world itself. And yet this naïve realist view that we can somehow perceive the world directly, is inconsistent with the physics of perception. If perception is a consequence of neural processing of the sensory input, a percept cannot in principle escape the confines of our head to appear in the world around us, any more than a computation in a digital computer can escape the confines of the computer.” [13] We process the data that we perceive; there is no ‘direct connection’ to the world around us.

Next let’s consider the concept of subjective constancy. Subjective constancy or perceptual constancy is the perception of an object or quality as constant under changing conditions.[14] In particular, subjective constancy in regards to color illustrates a human perceptual mechanism that can explain how some can perceive small differences, and some can’t. It is a matter of “need” and “interest.”

"The facts behind phenomena...are that we require fine color discriminations less frequently than gross discriminations, and when gross discriminations enable us to maintain focus on objects of prime interest, we 'systematically overlook' differences beyond the necessary degree of fineness. The mechanism which accomplishes this 'systematic overlooking' is the information-processing system of the organism, and the principle according to which it is accomplished is that this system never expands more of its capacity on a given perceptual task than is necessary according to the current needs and interests of the agent." [15]

“Just as in the case of the visual constancies, timbre constancy will have to be explained in terms of a complicated analysis by the brain, and not merely in terms of a simple registration of the input by the brain.” [16]

If the listener has a closed mind and has prejudged tonality to be a non-entity in comparison to technique, timing, and pitch accuracy, the “mechanism which accomplishes the 'systematic overlooking'” will be strongly in force, so that fine and even medium level tonal details will have a strong tendency to be overlooked. There will be no ‘need’ or ‘interest’ in perceiving details which are considered to be unimportant. It is an unfortunate fact that there are far too many musicians with impecable technique and poor tone. It will take practice and an open mind to allow today’s subtle details to become tomorrows glaring differences.

The difference in timbre between a bad instrument and a great instrument is huge; the difference between a very good instrument and a great instrument is comparatively small. How can a musician or instrument maker move the tone of an instrument in the direction of ‘greatness?’ First you have to be able to hear the differences, and then be able to define them. Once you know where in the tone the differences lie, then you can start to look for those things which will augment those tonal aspects you desire, and suppress those tonal aspects you dislike.

Wynton Marsalis once said that nuance is the highest order of musicality. The ability to notice fine detail levels, i.e. nuance, when addressing tonal issues is a necessity if you want to be able to make changes on that same level. Especially if you want to do so in a consistent and reliable fashion.

Successive papers will discuss the definitions and use of other objectively based concepts used in Tonespeak such as balance, center of tone, contrast, focus, rounded, sharp, and undertone as a means to further understanding and communication on the subject of tone quality.

A skill-building CD with examples of violin, viola, cello, and bass will soon (summer 2007) be available. Each string will be sonically highlighted so that all nine Tonespeak partitions are sonically examined and displayed.


        A Basic List of Some Commonly Used Tonal Descriptors:
A: airy
B: broad, beefy, buttery, bold, big, brassy, bright, brilliant, biting, buzzy, brittle, brash,
C: chocolate-y, creamy, centered, complex, clean, clear, crisp, cold, compressed, constricted
D: dull, dark, damped, dense, defined, detailed, dry, distorted, disconnected,
E: earthy, energetic, edgy
F: fat, full, focused, fast, fuzzy
G: gritty, grainy
H: heavy, hollow, hazy, harsh, hooty
I, J, K:
L: lumpy
M: mellow, muddy
N: narrow, nasal, natural
O:
P: pinched, punchy, percussive, penetrating, piercing,
Q: quick
R: restrained, resonant, rich, round, reedy,
S: soft, slow, sweet, strong, singing, silvery, sharp, strident, shrill,
T: tubby, thick, tight, transparent, thin, tinny
U, V:
W: warm, wide, wooly, woofy, woody, wet, weak, wispy, washed-out
X, Y, Z:


                     Some General Three Partition Interpretations

Too much energy in a partition; destructive contrast; out of balance:
Lows—muddy, tubby, thick, fat
Mids—nasal
Highs—harsh, hazy, shrill, thin, wispy, biting, cold, brash, strident, fuzzy, piercing

More energy in one or more partitions than others but in a good way; constructive contrast; has musically relevent character:
Lows—full, thick, big, fat
Mids—creamy, centered
Highs—clear, airy, bright, brilliant, silvery, percussive, defined
Mids & Lows—chocolate-y, full, mellow, warm, dense, thick
Mids & Highs—bold, singing

Not enough energy in a partition; destructive contrast; out of balance:
Lows—thin
Mids—hollow, dry
Highs—dull, soft, pinched


                        A Few General Nine Partition Interpretations

Brittle—having much more upper-highs than lower or mid-highs; too much contrast in the highs; often also not having sufficient support in the mids or lows
Centered—having a tonal balance that smoothly varies from one partition to another, one that has slightly more energy in the middle of the tone, and progressively slightly less energy as one moves away from the middle of the tone
Fuzzy—having an imbalance of noise (non-harmonically related energy) in the upper-highs
Fat—having more mid and upper-lows than lower-lows; good contrast; an overall balance that slightly to moderately favors the lows
Muddy—having more lower-lows than mid or upper-lows; too much contrast; an overall balance that at least slightly to moderately favors the lows



                                                  Definitions:

Balance
This term is used to indicate a positive quality assessment when a tone has it, and a negative quality assessment when a tone does not. The relative weighting of the overtones to the fundamental, and between the various overtones themselves, gives rise to this quality balance. A weighting formula that denotes good quality for one type of instrument will often be one that denotes poor or even bad tonal quality in a different type or family of instrument.

Big-ness
A “big” tone is one that has a balance of harmonics that favors the lower harmonics. It has slightly more lows than mids and highs, but preserves balance by not being too full in the lows to overshadow the mids and highs. A big tone with a slight imbalance (deficit) in the mids and highs can called various things such as heavy, or thick. The musical situation determines whether or not big-ness of tone needs some extra highs for contrast with the lows, or a lack of them for weight.

Center of tone
This term denotes a distribution of energy in a tone that at least slightly favors the middle part of that tone, in a rounded fashion. A quality tone should have a well developed ‘center of tone’ because it provides a richness and complexity to the tone that is appreciated by the listener.

Character
A term that describes a tone having two focuses, both towards the middle of the tone, but separate. This is a positive attribute: it has character.


Construct
A construct is a mental framework, a domain, which in our case has been created to deal with something perceived through one or more of our senses.

Contrast
Contrast is like texture in that one wants an interesting amount, but not too much. Tone has features, like contrast, big-ness (it’s a big tone), brightness, and smoothness. For example, if a tone has a wide rounded focus centered in the lower mids, a highlight in the lower or middle highs, if not too pointed, will add a nice contrast to a warm tone that might not project well. If the highlight is too prominent or pointed, it may be irritating rather than interesting. In this case there would be too much contrast.

Cross domain mapping
This is a cognitive process, a basic mechanism or way in which we seek understanding. It is the process of thinking about one thing in terms of another thing, without loosing sight of the characteristics of the first thing.

Experience
It’s often said that there is no substitute for experience. However, it’s an everyday thing to try to bypass experience. We all ask someone else what to do. We try to substitute some else’s experience for our lack of experience. That someone else may be a well known authority figure, a teacher, or someone ‘in the know.’ We want to get “the” answer on what to do to improve our tone. What reed/mouthpiece/strings should I use? That authority figure isn’t you, doesn’t have the same instrument as you do, isn’t at the same level of development as you. What works for them may not work at all for you. Learn to listen with as high a skill level as possible, as soon as possible. This way you can tell what your weak points are, and that will tell you where in your tone you need help. This in turn helps you to focus on the techniques or products that will actually help you with that problem area in your tone.

Focus
An area in a tone that attracts your attention because it is louder; it focuses your attention to one or more partitions. A focus may be good or bad, depending on where in the tone it resides (in which of the 9 partitions). If a focus is too low in the tone, it may unbalance a tone by making it sound too thick/slow. If a focus is too high in a tone, it may also cause tonal imbalance by making the tone appear thin or bright. A focus may be narrow or wide, rounded or sharp.

Graphic equalizer
A graphic equalizer is a stage, studio, and hi-fidelity audio control that allows the user to graphically see and control individually a number of frequency bands. A typical graphic equalizer consists of three or more electronic boost or cut tone controls, each having a different start and stop frequency in the audio band.

The volume control for each boost/cut control is usually a vertically oriented slide fader (a fader is also called a potentiometer or pot for short). A detent in the middle of the slide fader’s movement range signifies a no-change condition. If the fader knob is above this center position, it signifies that this part of the spectrum has been turned up. Conversely, if the fader knob has a position below this no-change center position, it signifies that this portion of the audio spectrum has been turned down.

The lowest portion of the audio spectrum that the graphic equalizer is designed to handle will be on the far left. Each succeeding band to the right is higher in frequency, and usually is in close proximity frequency-wise to the bands on either side.

Highlight
A highlight is the same as a focus.

Metaphor
This is a cross-domain mapping. The natural way we think is to conceptualize about one thing in terms of another.

Narrow
A narrow focus is one that is less than one partition wide.

Perceptual constancy
For instance, you can look at a house that is far away and still tell it is larger than the breadbox on your counter. You know it is large, even though it looks small, because you have mentally and automatically factored in it’s distance from you.

Pointed
A pointed focus has marked disparity in loudness across its bandwidth.

Round
A rounded focus has a smooth change in loudness across its bandwidth, with usually slightly more energy in the middle of the focus than at its extremes.

Spectral centroid
The spectral centroid of a tone is a single dividing line that creates two partitions in that tone’s bandwidth, each partition having equal energy. The higher in that tone’s bandwidth this dividing line is, the brighter this tone is perceived. It’s also called the ‘center of gravity’ of a tone. This is a mathematical/statistical term.

Textural descriptor
A textural descriptor is a word or phrase that describes a tonal quality or characteristic.

Timbre
A word that is used to talk about the amount and kind of quality the sound of a musical instrument contains. Timbre is functionally equivalent to the word tone.

Tone
A word that is used to talk about the amount and kind of quality the sound of a musical instrument contains. Tone is functionally equivalent to the word timbre.

Undertone
A band of energy in one or more partitions that has much less energy than the rest of the tone, but without which, the tone would be markedly changed, for good or bad.

Wide
A wide focus is one that extends across more than a single partition.


References:

1. from Wikipedia.org, on “pattern recognition”
2. Defining Timbre; J.Puterbaugh, http://www.music.princeton.edu/~john/definingtimbre.htm
3. Timbre Timeline; J. Puterbaugh, http://www.music.princeton.edu/~john/timbretimeline.htm
4. Timbre classification of a single musical instrument; M. Loureiro, H. de Paula, H. Yehia, 2004
5. The Timbre model; K. Jensen; Univ. of Copenhagen, Meeting of the A. S. A., 2002, p.1.
6. Spatial attribute identification and scaling by repertory grid technique and other methods; J. Berg, F. Rumsey, 1999, AES 16th International Conference
7. Salient feature extraction of musical instrument signals; T. Park, Masters thesis, Dartmouth College, 2000, p. 3.
8. The Contemporary Theory of Metaphor; G. Lakoff, 1992
9. Sonological models for timbre characterization; G. De Paoli & P. Prandoni, 1997
10. Perceptual effects of the spectral modifications of musical instruments; J. M. Grey & J. W. Gordon; Journal of the Acoustical Society of America, vol. 61, pp. 1270-1277.
11. Spectral centroid and timbre in complex, multiple instrumental textures; E. Schubert, J. Wolfe, A. Tarnopolsky; Proceedings of the 8th International Conference on Music Perception & Cognition, Evanston, IL, 2004.
12. Machine recognition of timbre using steady-state tone of acoustic musical instruments; I. Fujinaga, Peabody Conservatory of Music, Johns Hopkins University.
13. Gestalt isomorphism and the primacy of subjective conscious experience: a Gestalt bubble model; S. Lehar, Behavioral & Brain Sciences, vol. 26, no. 4, pp. 375-444, 2004.
14. from Wikipedia.org, on “subjective constancy”
15. Toward a Quantitative Model of Pattern Formation; Sayre, K.M. (1968); Philosophy and Cybernetics, p.149-152, ed. Frederick J. Crosson and Kenneth M. Sayre, Simon and Schuster.
16. Auditory Scene Analysis; Albert S. Bregman, MIT Press