What is timbre and can we measure it?

[solderdude]

There is a certain colouration that - even if due only to frequency variations - registers in my mind as a “ timber” placed across everything

Coloration ≠ timbre

I understand some people may want to call coloration 'timbre' but coloration is just coloration (a deviating frequency response).

 

[Timcognito]

Coloration ≠ timbre

I understand some people may want to call coloration 'timbre' but coloration is just coloration (a deviating frequency response).

Chromatics

 

[Axo1989]

Coloration ≠ timbre

I understand some people may want to call coloration 'timbre' but coloration is just coloration (a deviating frequency response).

Of course. I'd say colouration doesn't equal timbre, it's a component of it.

 

[Jim Taylor]

Of course. I'd say colouration doesn't equal timbre, it's a component of it.

Are we going to include every descriptive term in audio under the aegis of "timbre"?

Frequency response doesn't equal timbre, but it's a component of it.
Resonance doesn't equal timbre, but it's a component of it.
Harmonic distortion doesn't equal timbre, but it's a component of it.
Intermodulation distortion doesn't equal timbre, but it's a component of it.
Phase distortion doesn't equal timbre, but it's a component of it.
Beat frequencies don't equal timbre, but they're a component of it.

Where the hell is this going to end????????????????

 

[Axo1989]

Like I’ve mentioned before I also auditioned the audio physic Avanti. (And I’ve had the AP Virgos, Libras and Scorpios in my listening room). So I know what you mean.

I think it’s a fine line one might want to walk.

I personally wouldn’t want so much colouration that I give up on being informed about the nature of the recordings. Both because the distinct quality of recordings themselves are part of the fun, and also of course the recordings contain that timbrel information you don’t want to lose, so you want to make sure you don’t homogenize that away. ...

Apropos 'realism' I do enjoy the way my speakers do metallic sounds, and also evoke the sensations of drumsticks and hands hitting stretched skins, etc. I'll speculate that this comes in part from handling the onset/attack in a particular way (I could digress about their novel use of prosaic materials in cabinet and transducer decoupling roles and so on but that's straying a bit far, and quite speculative). More generally while I agree with other posters that overall envelope of natural instrumental sound/notess are longer duration than typical loudspeaker spectrogram decay, etc, I expect the onset/attack is also complex and may be very short duration. Better check.

On the vocal front, the crew building the house next door often play radio (ffs what century do they think this is?) but I rarely confuse the radio announcers' voices with the workers.

 

[Axo1989]

Are we going to include every descriptive term in audio under the aegis of "timbre"?

Frequency response doesn't equal timbre, but it's a component of it.
Resonance doesn't equal timbre, but it's a component of it.
Harmonic distortion doesn't equal timbre, but it's a component of it.
Intermodulation distortion doesn't equal timbre, but it's a component of it.
Phase distortion doesn't equal timbre, but it's a component of it.
Beat frequencies don't equal timbre, but they're a component of it.

Where the hell is this going to end????????????????

Haha, it's a sci-fi horror: Timbral Abyss. I always enjoy these movie threads.

*lost in the labyrinth of a 20-dimensional representation, is there no escape?
*yes I also started reading Acoustic Descriptors for Characterization of Musical Timbre Using the Fast Fourier Transform thanks to @Timcognito's reference query, not that I've digested this and other papers yet, but why not try a teaser:

4.1. Construction of the Objective Acoustic Parameter Set​

Timbre is a multidimensional perception attribute that is closely related to the time-domain waveform and spectral structure of sound [60]. In order to establish a timbre perception feature model, an objective acoustic parameter set was constructed using 54 parameters extracted from the timbre database. Objective acoustic parameters refer to any values acquired using a mathematical model representing a normal sound signal in the time and frequency domains. These 54 parameters can be divided into 6 categories [61]:
(1)
Temporal shape features: calculated from the waveform or the signal energy envelope (e.g., attack-time, temporal increase or decrease, and effective duration).
(2)
Temporal features: auto-correlation coefficients with a zero-crossing rate.
(3)
Energy features: referring to various energy content in the signal (i.e., global energy, harmonic energy, or noise energy).
(4)
Spectral shape features: calculated from the short-time Fourier transform (STFT) of the signal (e.g., centroid, spread, skewness, kurtosis, slope, roll-off frequency, or Mel-frequency cepstral coefficients).
(5)
Harmonic features: calculated using sinusoidal harmonic modeling of the signal (e.g., harmonic/noise ratio, odd-to-even and tristimulus harmonic energy ratio, and harmonic deviation).
(6)
Perceptual features: calculated using a model of human hearing (i.e., relative specific loudness, sharpness, and spread).

 

[MAB]

I've always thought of timbre as primarily the harmonic content. For instance, a clarinet playing a steady 440 Hz tone, same instrument but with two different embouchure produce two audibly different clarinet sounds:

First 20 harmonics provided for comparison. The harmonic content of a clarinet is evident, and the significant changes between the intensities of each harmonic between the two embouchure are also clear. In this example, I attempt to change only the volume of my throat and mouth while blowing the tone. For instance, the 7th harmonic changes by 20dB between the two embouchure! Each case still sounds distinctively like a clarinet, just different timbre. Other embouchure techniques, swapping mouthpiece, reed, and instrument all have even more dramatic changes. I consider all of this to be timbre, and other stylistic aspects like bending, attack, vibrato, staccato/legato, etc. to be different qualities of the sound than the timbre.

I have always read and heard timbre discussed as harmonic content, the couple references I have (like John S. Rigden's Physics and the Sound of Music page 151 for instance) define timbre as the harmonic frequencies and intensities and treat attack and decay of the tones as separate. I think the discussions of resonances in this thread are mostly the same as the the sources of harmonic content of a musical instrument. In fact, in the above example, the change in embouchure is just a change in the resonant cavity on one side of the reed.

Going with my working definition, a room or equipment can change the timbre by changing the relative amplitudes of the harmonics. Most of those changes will be somewhat haphazard, which is why some speakers sound so strange.

 

[Titurel]

Yes. An instrument maker cares about this a lot, of course. I tend to think of an instrument having potential only. But that is unmeasurable at rest. The sound when it is played has timbre, which is when we can measure it.

I think you touch on a fundamental tension in our competing world views. Musicians fall along a continuum from those who spend the bulk of their training and practice time learning to transcend their instruments and those who spend the bulk of their time trying to build the instrument nature gave them. Pianists tend towards the transcending end of things, singers, predictably, at the other end. A similar distribution, I believe, could be found between players of fixed pitched instruments (tempered) and those whose instruments are easily capable of playing in all keys (and alternative/novel tunings). I began my musical training in the tempered world, but spent the bulk of my academic and professional life in the (routinely*) untempered one.

* When you perform along with tempered instruments, you're kind of stuck living in their world.

 

[Titurel]

I've always thought of timbre as primarily the harmonic content. For instance, a clarinet playing a steady 440 Hz tone, same instrument but with two different embouchure produce two audibly different clarinet sounds:
View attachment 420608

First 20 harmonics provided for comparison. The harmonic content of a clarinet is evident, and the significant changes between the intensities of each harmonic between the two embouchure are also clear. In this example, I attempt to change only the volume of my throat and mouth while blowing the tone. For instance, the 7th harmonic changes by 20dB between the two embouchure! Each case still sounds distinctively like a clarinet, just different timbre. Other embouchure techniques, swapping mouthpiece, reed, and instrument all have even more dramatic changes. I consider all of this to be timbre, and other stylistic aspects like bending, attack, vibrato, staccato/legato, etc. to be different qualities of the sound than the timbre.

I have always read and heard timbre discussed as harmonic content, the couple references I have (like John S. Rigden's Physics and the Sound of Music page 151 for instance) define timbre as the harmonic frequencies and intensities and treat attack and decay of the tones as separate. I think the discussions of resonances in this thread are mostly the same as the the sources of harmonic content of a musical instrument. In fact, in the above example, the change in embouchure is just a change in the resonant cavity on one side of the reed.

Going with my working definition, a room or equipment can change the timbre by changing the relative amplitudes of the harmonics. Most of those changes will be somewhat haphazard, which is why some speakers sound so strange.

Absolutely. The reed in a clarinet (oboe, et al) is also affected by its back-loading, the oropharynx, as well as variations of the embouchure.

 

[MattHooper]

I've always thought of timbre as primarily the harmonic content. For instance, a clarinet playing a steady 440 Hz tone, same instrument but with two different embouchure produce two audibly different clarinet sounds:
View attachment 420608

First 20 harmonics provided for comparison. The harmonic content of a clarinet is evident, and the significant changes between the intensities of each harmonic between the two embouchure are also clear. In this example, I attempt to change only the volume of my throat and mouth while blowing the tone. For instance, the 7th harmonic changes by 20dB between the two embouchure! Each case still sounds distinctively like a clarinet, just different timbre. Other embouchure techniques, swapping mouthpiece, reed, and instrument all have even more dramatic changes. I consider all of this to be timbre, and other stylistic aspects like bending, attack, vibrato, staccato/legato, etc. to be different qualities of the sound than the timbre.

I have always read and heard timbre discussed as harmonic content, the couple references I have (like John S. Rigden's Physics and the Sound of Music page 151 for instance) define timbre as the harmonic frequencies and intensities and treat attack and decay of the tones as separate. I think the discussions of resonances in this thread are mostly the same as the the sources of harmonic content of a musical instrument. In fact, in the above example, the change in embouchure is just a change in the resonant cavity on one side of the reed.

Going with my working definition, a room or equipment can change the timbre by changing the relative amplitudes of the harmonics. Most of those changes will be somewhat haphazard, which is why some speakers sound so strange.

Nice contribution!!

 

[Axo1989]

Chromatics

Where the hell is this going to end????????????????

 

[Galliardist]

While I slept, some of you brought up questions about performer effects on timbre or the role of performer in timbre. These arguments have merit, primarily because they focus our attention on the actual production of the sound(s) on to which timbral descriptions might be hung.

For the sake of argument, allow me to temporarily adopt a combined timbre/envelope view of timbre. I use violin only because it provides an easy case study. Most other instruments are also "timbre malleable" depending on the means of excitation chosen. N.b., timbre does depends on manner of excitation, not on envelope, which is a product of the same choice of means.

When someone talks about violin timbre to a violinist, the violinist could easily ask” WHICH timbre were you talking about: the standard bowed timbre (arco), or pizzicato, or col legno. etc.?” There are dozens of conventional techniques for squeezing sound (and thus timbre) out of a violin, each with it’s own characteristic timbre and envelope. Yet the average listener to classical music would recognize the violin’s timbre with all but the most atypical performing techniques. In this context, conflating timbre and envelope is an unnecessary complication, which is why I argue for envelope as an independent, but closely coupled aspect of musical sound; because it is.

In simple terms, the player controls sound production, consequently affecting timbre and envelope independently if in closely associated ways. Coincidence or correlation is not proof of causality (added: or entanglement).

The violin techniques mentioned above can all be notated by the composer, conductor, section leader, or individual player. To some degree they modify the characteristic timbre and define (or constrain) the envelope. The degree to which these choices apply to sampled or synthesizsed sound depends on the depth (and ready availability) of envelope parameters in the player interface. Some of the physical modeling schemes are quite capable of convincing envelope analogs, but only to the extent the player interface allows.

Responding here to a recent post regarding piano and guitar; The two instruments (the piano especially) almost give credence to the ADSR model, but they are anomalous instruments (again the "unprepared" piano, particularly). Unfortunately, they are favorite instruments of theorists and academics, who are prone to treat them as representative, rather than as well populated ghettos in the diverse world of music instruments. Theories which rely too heavily on them are immediately suspect.

Interestingly, I found a different perspective on this from a session demonstrating different classical guitar constructions I attended several years ago, where we were told that most academic work on issues of sound production are carried out on "period" and orchestral instruments, though the guitar and particularly the classical guitar was producing a lot of material on construction methods and measurement techniques because there is a wider range of construction methods and string materials in use than for other "conservatory" instruments.

(I've not looked to see if this is the case, since I live as an amateur classical guitarist in one of your "well populated ghettos" and I got dragged into the other one by one of my partner's writing projects a few years ago. I'm not an academic or theorist.)

A point made by a guitarist at that session was that in the case of the guitar, at least, the instrument decides the final timbre and available dynamics (classical guitar doesn't have a proper forte compared to most instruments) when playing, whether using standard or extended techniques - "good technique gives the instrument a better chance of sounding optimally". The player then uses that timbre - rather, those timbres, and the available dynamic, to construct the performance and has control within the limits of the instrument.

I do note that most instruments seem to have "concert" or "professional" versions that improving players upgrade to, which tells me that the instrument plays a role in the final result more generally, though I would suspect from your posts that the exact role and relationship may change from instrument to instrument.

The post I made that mentioned piano and guitar though (which I presume you are referring to), was not about a theory of timbre, though. Go back and read it. It was about reproduction. Does the timbre of your playing and instrument survive similar downgrades in sound quality and remain recognisable? I bet it does, but would be interested to know your take on that.

 

[Titurel]

The post I made that mentioned piano and guitar though (which I presume you are referring to), was not about a theory of timbre, though. Go back and read it. It was about reproduction. Does the timbre of your playing and instrument survive similar downgrades in sound quality and remain recognisable? I bet it does, but would be interested to know your take on that.

Actually, I don't think I was responding to your post in specifically. Rather, I noticed several posts mentioning guitars and pianos and this spurred me to post an observation that theorists tended to be keyboard-centric, and that such youtube pondefecators as messers Beato and Neely live in guitar-centric worlds. That said...

I did go back and read a number of your posts and I find that we are, if not always on the same page, at least in the same chapter. As to your question regarding reproduction: downgrades to the reproduction chain inevitably degrades timbral integrity and, eventually, identity. How fatal the degradation is depends on one's ability to mentally fill in the gaps. With early electrically miked operatic recordings, I can usually hear past the distortions and enjoy the performance. Acoustic era recordings are harder work.

I have a small collection of historical recordings, e.g., Alesandro Moreschi, the only solo recorded castrato connected to bel canto era. With his and similar recordings, I believe I can almost reconstruct (in my head) what the original sound was like, but only because I'm a fair dinkum vocal pedagogue. I can often hear what he's trying to do, even when I can't hear how well he succeeds. I doubt I'd have as much confidence listening to similarly ancient saxophone recordings.

 

[IPunchCholla]

I assumed the reason for using piano in particular as a theoretical starting point was that it was simpler in terms of demonstrating ADSR than say reed or brass instruments. Guitar because again it is simpler than bowed instruments in terms of ADSR. Therefore they are easier to demonstrate the principles on not that they are unique or that other instruments don’t follow the same rules.

Basically piano is western music theory’s spherical cow.

 

[gnarly]

To anyone who thinks timber would only apply to individual instruments, I’d ask:

Is there any good non-arbitrary reason to restrict this to single instruments and not to more than one instrument?

For 60 years, every time I've heard the word timbre, it has been in association with an individual acoustic source.
Never in association with a collection of acoustic sources.

I see historical use supporting the idea that timbre is a descriptor of the difference in the sound of individual acoustic sources creating the same note.
I think if we extend timbre to describe the difference in the sound of summed multiple sources, it's lost all of its meaning.

I see further in this thread you search for speakers that you feel accurately portray the timbe of instruments and vocalists. So do I.
I'm 100% convinced speakers can do nothing but detract from the timbre heard from real individual acoustic sources.
The goal for me is finding speakers that detract from timbre as little as possible.
IOW, I rate the best speakers as having no influence on timbre at all. They simply let me hear more of its nuances, by being less destructive.

 

[bothu]

Where the hell is this going to end????????????????

You made my day ..... I bow my bold head to you !

 

[Titurel]

I assumed the reason for using piano in particular as a theoretical starting point was that it was simpler in terms of demonstrating ADSR than say reed or brass instruments. Guitar because again it is simpler than bowed instruments in terms of ADSR. Therefore they are easier to demonstrate the principles on not that they are unique or that other instruments don’t follow the same rules.

Basically piano is western music theory’s spherical cow.

Simple yes, representative no. MOO indeed. While the piano, in addition to its role as solo instrument, is a fair stand in for the orchestra, it is a poor stand in for any individual orchestral instrument except for those from in own instrument family - percussion. The piano has much more in common with the timpani, than it does with the oboe, though, in terms of pitch malleability, even the timpani is better than the piano. The guitar has similar problems, but it is more malleable than the piano in both the pitch and envelope domains, primarily because its player has nearly unlimited direct access to the strings. Because players of the piano and guitar have severely limited control of one or more crucial parameters, those instruments make poor test beds for the study of timbre generally.

Synthesizers, depending on their architectures and player interfaces, have their own problems as test beds. Rather than deal with them all, let's look at sample-based synthesizers. Samples, for a number of very good reasons*, usually have little or no room signature, which is then added (synthesized) after the fact. Thus, to make a sample more generally useful, the sampler omits (or limits) a crucial timbral parameter - the instrument's acoustical output impedance.

Not all instruments couple to the room identically. Run a spinorama on a variety of instruments and you'll see what I'm saying. If you omit or limit room signature from consideration, you're not examining timbre. Compare the French Horn and Mellophone, two morphologically similar instruments. The Mellophone's sound is directional, radiating forward of the player. The direct sound dominates. The French Horn (usually) points backwards, its output variably occluded by the player's right hand. Its timbre is dominated by reflected sound. Stripped out of their usual environments, the two instruments' sounds would be predictably difficult to differentiate, even if you include ADSR. MOO! Now consider the acoustical radiation of all the other instruments, starting with the violin, tuba, flute, harp, human voice.

(Does anyone have a cow emoji?)


* an interesting topic all by itself

 

[IPunchCholla]

Simple yes, representative no. MOO indeed. While the piano, in addition to its role as solo instrument, is a fair stand in for the orchestra, it is a poor stand in for any individual orchestral instrument except for those from in own instrument family - percussion. The piano has much more in common with the timpani, than it does with the oboe, though, in terms of pitch malleability, even the timpani is better than the piano. The guitar has similar problems, but it is more malleable than the piano in both the pitch and envelope domains, primarily because its player has nearly unlimited direct access to the strings. Because players of the piano and guitar have severely limited control of one or more crucial parameters, those instruments make poor test beds for the study of timbre generally.

Synthesizers, depending on their architectures and player interfaces, have their own problems as test beds. Rather than deal with them all, let's look at sample-based synthesizers. Samples, for a number of very good reasons*, usually have little or no room signature, which is then added (synthesized) after the fact. Thus, to make a sample more generally useful, the sampler omits (or limits) a crucial timbral parameter - the instrument's acoustical output impedance.

Not all instruments couple to the room identically. Run a spinorama on a variety of instruments and you'll see what I'm saying. If you omit or limit room signature from consideration, you're not examining timbre. Compare the French Horn and Mellophone, two morphologically similar instruments. The Mellophone's sound is directional, radiating forward of the player. The direct sound dominates. The French Horn (usually) points backwards, its output variably occluded by the player's right hand. Its timbre is dominated by reflected sound. Stripped out of their usual environments, the two instruments' sounds would be predictably difficult to differentiate, even if you include ADSR. MOO! Now consider the acoustical radiation of all the other instruments, starting with the violin, tuba, flute, harp, human voice.

(Does anyone have a cow emoji?)


* an interesting topic all by itself

Room coupling is an interesting consideration that I hadn’t thought about. But given that musicians at least claim to be able to identify exemplars of the same instrument even when played in different locations, room coupling itself must be heard through to some extent, making me wonder if room coupling should be considered timbre, as it is a part of the instrument class (in addition to being heard through) not a differentiator of the instrument.

another way to say it is that timbre is ADSR and sonic spectrum of an instrument. Room coupling is (a minor) part of how that sonic spectrum is formed.

 

[MattHooper]

I think if we extend timbre to describe the difference in the sound of summed multiple sources, it's lost all of its meaning.

I don’t see how that’s the case. That seems like an unnecessary slippery slope argument.

Timbre is the phenomenon by which we can distinguish different sounds, through features like different harmonics, and overtone combinations, attack, and decay, resonance, etc.

“Timbre” is often used in describing musical instruments yes. But it’s a set of real world phenomena that is hardly confined to instruments.

If I have a drumstick and I tap my wooden kitchen table, and then I tap a metal pan, we will be able to differentiate between those sounds. Why? It will be due to largely to the difference in Timbre - differences in harmonics and overtone combinations, attack, and decay, resonance,

It’s the same phenomenon, so why would we use a different word for it?

Likewise, if you compare the sound of a single violin playing a note versus a mass of violins playing the same note (even at the same volume ) how will you tell them apart?

If it’s not by the same qualities known under the term Timbre?

It’s almost like saying that we shouldn’t use the same term “red” to describe both a car painted red and a flower that is red. Because that would be somehow confusing.
Why not? You can understand they are different things and yet identify the phenomenon of “ red colour” also applies to each.

I don’t see how it’s confusing at all If you simply identify what you were talking about.
On one hand, I’m talking about the timber of a single violin note and on the other hand I’m clear about talking about the timber of massed strings.

 

[Titurel]

Room coupling is an interesting consideration that I hadn’t thought about. But given that musicians at least claim to be able to identify exemplars of the same instrument even when played in different locations, room coupling itself must be heard through to some extent, making me wonder if room coupling should be considered timbre, as it is a part of the instrument class (in addition to being heard through) not a differentiator of the instrument.

Without "coupling" you don't hear anything. The most desired requirement in room coupling would be that whatever coupling there is doesn't degrade a signal's timbral signature to the point of where the timbre is unrecognizable or ambiguous. If we could think of timbre like (vowel) formants, identifying its significant components might then become more tractable.. (Familiarity with the "singer's formant" might help, here.)

Room coupling is (a minor) part of how that sonic spectrum is formed.

I would recast the above as "room coupling is an unavoidable (potentially minor) part of how the sonic spectrum is formed." (N.b., room might be unbounded - outdoors; also notice ADRS's absence.) Going on, room coupling only becomes an issue when the inherent resonances interfere with crucial formants. My media room is a good example. Prior to installing clouds over the LP, speech was difficult to understand, owing to cancelations - a product of LP, speaker position, and room geometry. To make this problem description "timbral," one need only parameterize the positions and room geometry. That is, a room has timbre, but its effect (on reproduction) is mostly felt where potentially interfering (or disruptive) resonances intersect with source (speaker) and/or receiver (listener) positions. With a spinorama (or similar) derived timbral model, timbre becomes 3d and it gains specificity. This would mesh well with a general model of timbre that recognized the importance of dispersion (directivity) to timbre.

Like I said earlier, timbre is a lot more complicated than it appears, and simple metrics probably won't crack its nut.