Fight for a sound: timbre - psychoacoustics vs subjectivism
Among people who, by virtue of their hobbies and (or) professional duties, are faced with an assessment of the quality of audio equipment , two categories can be distinguished. Those who prefer to judge quality on the basis of objective criteria (comparing data with accepted standards) and those who believe that sound quality should be determined by hearing and believe that the value of objective indicators is exaggerated.
A large number of articles about the quality and features of sound, as well as comments and disputes in these articles on the GT, demonstrates the genuine interest of users to the topic. I doubt that this article will stop the controversy of adherents of one and the second point of view, but perhaps it will shed a ray of light in the “pitch darkness” of quality assessment issues.
Due to the fact that quite a lot has been written about the classical criteria for evaluating amplifiers, acoustics and other things, for example, the recent article by Oleg Shulgin (I highly recommend), in this material I will focus readers on the possibilities of objective measurement of timbre sound characteristics.
As a rule, supporters of subjectivism base the “correctness” of their judgments on statements like: “In sound, not everything can be measured, therefore, it should be evaluated solely with our ears and nothing else. Objective data gives me nothing as a listener. ” In most cases, the stumbling block becomes the problem of assessing the timbre features of the sound of the equipment.
')
Often, such people write and say that the timbre is very difficult for hardware evaluation, does not reflect the real picture of what they mean by the character of the sound. If we ask them how they describe these features, they will immediately reply that the sound should be “soft” (“hard”, “fluffy”), airy (“not soapy”, “dense”, “discharged”) , “Dry”, “wet”, etc.
At first glance, it is difficult to refute or confirm their statements, because both the descriptive terminological apparatus and the assessment are subjective, however, this is not quite so. The fact is that very many “extremely subjective” characteristics, descriptive terms, and so forth. The effects have quite understandable, measurable reasons, and most of the factors affecting timbre are studied in detail.
One of the main tasks of psychoacoustics, as a related branch of physical and psychological knowledge, is the search for correspondences between objective physical characteristics and its subjective assessment. The timbral features are no exception, and despite the complexity of the task, they are studied quite well.
Studies of timbre as an acoustic phenomenon have been conducted for 200 years. The factors that form the timbre have already been found, and each of these factors has clear, measurable characteristics. In identifying objective indicators of timbre, such scientists as Helmholtz (1877), Fletcher (1938), Liklide (1951), Plom (1976), Noutsm (1989), Rossing (1990), Hande (1995) made a contribution.
In accordance with the American standard ANSI-60, timbre is defined as:
Most researchers understand the timbre as “tone quality” and “color tone”. To describe the timbre, supporters of the subjective approach use such terms as: dry, sonorous, soft, sharp, bright and warm, cold, beautiful, which, at first glance, cannot reflect the objective parameters.
Helmholtz in the definition, ahead of his time, gave the following timbre characteristic:
While conducting his experiments, the scientist also made several observations, which are fully confirmed by modern research. For example, it is in the works of Helmholtz that information about the dependence of the perception of a timbre on the speed of “entry” of partial tones at the beginning of a sound and their “death” at the end first occurs. It is also this pioneer of acoustics that linked the perception of timbre (recognition of timbres of individual instruments) with the presence of certain noises and irregularities.
H. Fletcher
In 1938, Fletcher proved that the timbre depends on the structure of the overtones and can vary significantly depending on the volume and pitch. In 1976, Plomp experimentally confirmed that human hearing does not suffer from “phase deafness,” and accordingly, the perception of timbre depends both on the shape of the spectral envelope of the amplitude spectrum, and on the spectrum of the phases. In 1990, Rossing added that the timbre depends on the temporal envelope of the sound and its duration. As can be seen from the history of studies of timbre, scientists have fully proved the possibility of physical measurement of the parameters of timbre coloring.
In accordance with research, the main physical factors that determine the timbre of a sound are:
General view of the spectral envelopes and their change over time
As a basis for subsequent correlation studies, a standardized table of subjective descriptions of timbres was created, most of which were included in the AES-20-96 standard - “Recommendations for the subjective assessment of loudspeakers”. It is from here that some marketers draw definitions for speculation.
In the process of research, scientists reveal a strict correspondence of each of the descriptions to exact physical parameters.
Due to the complex influence of factors shaping the timbre, there are no common units of measurement. This does not mean the impossibility of objective measurement and obtaining accurate data. Currently, Stanford University, Harvard Laboratories, IRKAM and Berkeley University are working to create exact correspondences between the subjective descriptive terms and the objective characteristics of factors influencing the formation of timbre.
LA study Kuznetsov to create a reference book "Acoustics of musical instruments" (the conclusions of the study are fully valid for audio equipment ) gave results for several characteristics of the timbre. Thanks to the work of the scientist, it became known that the main parameter for these characteristics is the envelope of the amplitude spectrum. We can briefly present Kuznetsov’s conclusions as follows:
Further experiments with the evaluation of the timbre characteristics of the sound of the acoustic systems made it possible to establish more accurate correlation correspondences between the peaks of the frequency response and subjective changes in the tone.
From the above we can make a simple conclusion: in physics and psychoacoustics for timbre there are precise definitions, and any “airy”, “dry” or notorious “warm tube” sound is determined by specific physical parameters that have nothing to do with the metaphysical concepts of some audiophiles. Until the last strike on marketing myths, it did not remain long, since, for example, the AES-20-96 standard - “Recommendations for the subjective assessment of loudspeakers”, in light of research on the nature of timbre, ceases to be relevant. Hande's 1995 study “On the effect of changing the shape of the spectral envelope and the nature of its changes over time on timbre” can be considered a point of reference, after which any statement about the incomprehensibility of the nature of “airiness”, “openness” or “transparency” of sound looks ridiculous.
A large number of articles about the quality and features of sound, as well as comments and disputes in these articles on the GT, demonstrates the genuine interest of users to the topic. I doubt that this article will stop the controversy of adherents of one and the second point of view, but perhaps it will shed a ray of light in the “pitch darkness” of quality assessment issues.
Due to the fact that quite a lot has been written about the classical criteria for evaluating amplifiers, acoustics and other things, for example, the recent article by Oleg Shulgin (I highly recommend), in this material I will focus readers on the possibilities of objective measurement of timbre sound characteristics.
Sound problem
As a rule, supporters of subjectivism base the “correctness” of their judgments on statements like: “In sound, not everything can be measured, therefore, it should be evaluated solely with our ears and nothing else. Objective data gives me nothing as a listener. ” In most cases, the stumbling block becomes the problem of assessing the timbre features of the sound of the equipment.
')
Often, such people write and say that the timbre is very difficult for hardware evaluation, does not reflect the real picture of what they mean by the character of the sound. If we ask them how they describe these features, they will immediately reply that the sound should be “soft” (“hard”, “fluffy”), airy (“not soapy”, “dense”, “discharged”) , “Dry”, “wet”, etc.
At first glance, it is difficult to refute or confirm their statements, because both the descriptive terminological apparatus and the assessment are subjective, however, this is not quite so. The fact is that very many “extremely subjective” characteristics, descriptive terms, and so forth. The effects have quite understandable, measurable reasons, and most of the factors affecting timbre are studied in detail.
The study of timbre in psychoacoustics
One of the main tasks of psychoacoustics, as a related branch of physical and psychological knowledge, is the search for correspondences between objective physical characteristics and its subjective assessment. The timbral features are no exception, and despite the complexity of the task, they are studied quite well.
Researcher Irina Aldoshina about the timbre: “There are difficulties with the definition of this term that are comparable to the definition of the term“ life ”: everyone understands what it is, but science has been beating on the scientific definition for several centuries.”
Studies of timbre as an acoustic phenomenon have been conducted for 200 years. The factors that form the timbre have already been found, and each of these factors has clear, measurable characteristics. In identifying objective indicators of timbre, such scientists as Helmholtz (1877), Fletcher (1938), Liklide (1951), Plom (1976), Noutsm (1989), Rossing (1990), Hande (1995) made a contribution.
In accordance with the American standard ANSI-60, timbre is defined as:
“An attribute of aural perception, which allows the listener to judge that two sounds having the same pitch and volume are different from each other.”
Most researchers understand the timbre as “tone quality” and “color tone”. To describe the timbre, supporters of the subjective approach use such terms as: dry, sonorous, soft, sharp, bright and warm, cold, beautiful, which, at first glance, cannot reflect the objective parameters.
Helmholtz in the definition, ahead of his time, gave the following timbre characteristic:
“The difference in the musical quality of the tone (timbre) depends only on the presence and strength of the partial tones (overtones), and does not depend on the phase difference with which these partial tones come into the composition”
While conducting his experiments, the scientist also made several observations, which are fully confirmed by modern research. For example, it is in the works of Helmholtz that information about the dependence of the perception of a timbre on the speed of “entry” of partial tones at the beginning of a sound and their “death” at the end first occurs. It is also this pioneer of acoustics that linked the perception of timbre (recognition of timbres of individual instruments) with the presence of certain noises and irregularities.
H. Fletcher
In 1938, Fletcher proved that the timbre depends on the structure of the overtones and can vary significantly depending on the volume and pitch. In 1976, Plomp experimentally confirmed that human hearing does not suffer from “phase deafness,” and accordingly, the perception of timbre depends both on the shape of the spectral envelope of the amplitude spectrum, and on the spectrum of the phases. In 1990, Rossing added that the timbre depends on the temporal envelope of the sound and its duration. As can be seen from the history of studies of timbre, scientists have fully proved the possibility of physical measurement of the parameters of timbre coloring.
In accordance with research, the main physical factors that determine the timbre of a sound are:
- lining up the amplitudes of overtones during the attack;
- change of phase relations between overtones from conditioned to random ones;
- changing the shape of the spectral envelope over time during all periods of sound development (attack, stationary part and attenuation);
- the presence of irregularities of the spectral envelope and the position of the spectral centroid (maximum spectral energy, which is associated with the perception of formants) and their change over time;
- the presence of modulations - amplitude (tremolo) and frequency (vibrato) characteristics;
- changing the shape of the spectral envelope and the nature of its change over time;
- change in intensity (volume) of sound, i.e. the nature of the nonlinearity of the sound source;
- the presence of additional signs of identification of sources (instrument, apparatus).
General view of the spectral envelopes and their change over time
You thought that "velvety" can not be measured - you were wrong
As a basis for subsequent correlation studies, a standardized table of subjective descriptions of timbres was created, most of which were included in the AES-20-96 standard - “Recommendations for the subjective assessment of loudspeakers”. It is from here that some marketers draw definitions for speculation.
In the process of research, scientists reveal a strict correspondence of each of the descriptions to exact physical parameters.
Due to the complex influence of factors shaping the timbre, there are no common units of measurement. This does not mean the impossibility of objective measurement and obtaining accurate data. Currently, Stanford University, Harvard Laboratories, IRKAM and Berkeley University are working to create exact correspondences between the subjective descriptive terms and the objective characteristics of factors influencing the formation of timbre.
LA study Kuznetsov to create a reference book "Acoustics of musical instruments" (the conclusions of the study are fully valid for audio equipment ) gave results for several characteristics of the timbre. Thanks to the work of the scientist, it became known that the main parameter for these characteristics is the envelope of the amplitude spectrum. We can briefly present Kuznetsov’s conclusions as follows:
- a smooth rise of the envelope (an increase in the amplitudes of a certain group of overtones) in the region of 200 ... 700 Hz makes it possible to obtain shades of richness and depth;
- the rise in the range of 2.5 ... 3 kHz gives timbre "flight", "sonority";
- the rise in the range of 3 ... 4.5 kHz gives a timbre "sharpness", "piercing", etc.
Further experiments with the evaluation of the timbre characteristics of the sound of the acoustic systems made it possible to establish more accurate correlation correspondences between the peaks of the frequency response and subjective changes in the tone.
Total
From the above we can make a simple conclusion: in physics and psychoacoustics for timbre there are precise definitions, and any “airy”, “dry” or notorious “warm tube” sound is determined by specific physical parameters that have nothing to do with the metaphysical concepts of some audiophiles. Until the last strike on marketing myths, it did not remain long, since, for example, the AES-20-96 standard - “Recommendations for the subjective assessment of loudspeakers”, in light of research on the nature of timbre, ceases to be relevant. Hande's 1995 study “On the effect of changing the shape of the spectral envelope and the nature of its changes over time on timbre” can be considered a point of reference, after which any statement about the incomprehensibility of the nature of “airiness”, “openness” or “transparency” of sound looks ridiculous.
Source: https://habr.com/ru/post/399953/
All Articles