Singing With An ‘Open Throat’: Vocal Tract Shaping
(Page 3 of 3)
The study of formants, or at least the acquisition of a basic understanding of them, is a vital part of vocal training, and should not be neglected by serious singers wishing to produce the absolute best vocals that they possibly can. The ‘ring’ or the ‘focus’ of the voice, which is reflective of balanced resonance, depends on the presence of acoustic strength in the upper regions of the spectrum. In other words, it can only be achieved if upper partials or overtones (formants) are present. How to encourage the presence of these overtones, and thus positively affect the quality of the tone produced, is an important skill for all serious singers to gain.
In the following sections, I will discuss the relationship between vocal tract shaping and tone – how correct articulation of vowels creates balanced, fully resonant tone that is marked by the presence of these vocal formants.
Voiced sounds are acoustically rich, having many harmonics above the fundamental frequency (the lowest frequency of a complex sound, which corresponds to the unique pitch heard in such a complex tone). These harmonics or overtones are integer or whole number multiples of the fundamental frequency, and occur at roughly 1000Hz intervals. (Spectrograms are often used to visualise and track formants.) Put very simply, the complex sound of the voice resonates at different harmonic pitches. These resonance frequencies, each corresponding to a resonance in the vocal tract - the ‘pipe’ between the ‘voice box’ (larynx) and the mouth - are called formants.
Because of their resonant origin, formants tend to stay essentially the same even when the frequency (pitch) of the fundamental is changed continually. As I will explain in an upcoming article on vowels and vowel modification, all vowels have their own unique formant frequencies – they are defined by their distinct frequencies at the first and second formants - that don’t change even when pitch changes. For example, the formant frequencies for the [i] vowel for any given voice are more or less constant and remain within very specific limits in the frequency range. For this reason, these vowel formants may be called ‘fixed formants’. (These frequencies remain constant because the articulation of the individual vowels remains relatively the same regardless of pitch.)
The spaces above the vocal folds are a series of connected resonating chambers that filter the sounds that emanate from the voice source (the vocal folds inside the larynx). During speech production and singing, the source signal – the sound wave produced by the larynx - is filtered according to the morphology of the oral tract and of the articulators (e.g., tongue, jaw, lips, etc.). Therefore, formants are acoustic resonances of the vocal tract itself that result from the various dimensions of the vocal tract spaces.
Because the vocal tract is a complex acoustic tube resonator of varying sizes and shapes, and is highly adjustable – it’s not a simple tube – it tends to emphasize and amplify some overtones (harmonic components) of the phonated sound and de-emphasize or dampen others. (The resonance frequencies of the vocal tract tend to emphasize a series of frequencies that relate to the size and shape of the vocal tract, although these frequencies are not sequential in their strength, as in the case with overtones of the phonated sound.)
The vocal tract resonator has different requirements for the sounds that try to pass through it, depending upon the frequency of that sound. Certain frequencies pass through the resonator easily and, as a consequence, are given a high amplitude. Because partials of various frequencies are transmitted through the vocal tract simultaneously, those that coincide with the formant frequencies are radiated from the lip opening (projected from the mouth) with greater strength than others. In other words, harmonics that fall at or near these resonance frequencies of the vocal tract pass freely through the vocal tract and are most efficiently radiated as sound, producing a formant. Therefore, formants appear as peaks in the spectrum of the radiated sound. (Harmonics whose frequencies are not close to the resonance frequencies of the vocal tract become weakened, forming troughs between the spectral peaks, and do not pass through the vocal tract.)
Some of the parameters involved in the filtering phenomenon that brings about the formant pattern are speaker specific. For example, the length of the pharynx and the size of the vocal chamber distinguish the timbre of women, men and children. (Adult females have shorter vocal tracts than adult males. Therefore their formant frequencies are fifteen percent higher on average than those of the adult male.)
The timbre of the voice strongly depends on the formant pattern (but it is also influenced by the signal's pitch), in particular on the first three formants. However, the same speaker can modulate a wide repertoire of sounds by changing the morphology of the vocal cavities, namely by displacing the tongue, opening the oral chamber to different extents, increasing its size by rounding and protruding the lips, letting the sound wave pass through the nasal cavity (instead of the oral one), or by lowering the velum (soft palate). All these parameters affect the formant pattern of the sound's spectrum, and the formant pattern enables us to distinguish between voiced sounds, in particular between oral vowels - an ordinary vowel without nasalisation.
Formants determine vowel quality and donate personal timbre to the voice. As well as determining perceived vocal quality, formants are important for the perception of personal voice quality, permitting us to identify individual voices, such as those of favourite singers on CD recordings or family members on the telephone.
In speech communication, our perception and labeling of certain sounds as vowels depends on the relationship of the lowest two formants. (In other words, the first two formants strongly contribute to the differentiation of vowels from one another.) For example, the ‘roundness’ or ‘depth’ – the ‘oscuro’ factor of the professional chiaroscuro timbre - of the sound results from the presence of strength in lower partials of the spectrum, termed the first formant.
Vowel ‘colour’ is also determined by the two lowest formants. It is defined mainly by the location in the spectrum of the changing middle formant – the second formant.
Timbre is determined by the third, fourth, and fifth formants.
When considering the different sounds produced during speech, usually just the first, second and third formants are considered, since these are the only formants whose frequencies tend to vary, (generally according to the vowel being sung).
There are generally five formants relevant to singing, though, and they are crucial in the perception and discrimination of voiced sound. In singing, contributions to the overall projection of sound are believed to be made by formants higher than the third. In fact, six or seven formants can often be identified in the laboratory. The higher formants are thought to contribute to the individual identity of the speaking or singing voice.
Swedish physician and medical researcher Johan Sundberg has identified an additional concentration or prominent cluster of intense acoustic energy, consisting of strong third, fourth and fifth formants. This cluster, which he termed the Singer’s Formant, results from the cumulative distribution of upper harmonic partials (or overtones) that is present in the frequency spectra only of trained singing voices. This formant, which seems to be independent of the particular vowel and pitch, adds brilliance and carrying power to the voice. Tonal balance – or chiaroscuro timbre - is enhanced by, and results from, a proper distribution of energy among these three formant regions.
In the trained singing voice, considerable acoustic strength is present in both upper and lower regions of the spectrum regardless of the vowel being sung. The proper balance of this acoustic energy in upper, middle and lower portions of the spectrum ensures the classical resonance balance of the singing voice. The combining of these first two formants along with the cluster of formants – the third, fourth and fifth formants (generally referred to as the Singer’s Formant that is found in some ranges of the singing voice in close proximity to the third formant) produces the ideal clear/dark tone of the historic chiaroscuro singing timbre. While the third formant - the Singer’s Formant - produces the chiaro (light or clear) aspect of the historic chiaroscuro tone of the singing voice, it is largely the first formant that produces the balancing oscuro (dark) aspects.
A study on developing good tone is not complete without a look at how even small adjustments made to the vocal tract can change the overall tone quality of the singing voice by changing the relationship of the formants to the fundamental frequency.
The vocal tract is capable of being shaped in a variety of ways, and how it is shaped directly affects the quality of the tone produced by the singer. Any change in the cross section of the vocal tract shifts the individual formant frequencies, the direction of the shift depending on just where the change in area falls along the standing wave. For example, constriction of the vocal tract at a place where the standing wave of a formant exhibits minimum amplitude pressure oscillations generally causes the formant to drop in frequency. Expansion of the tract at those same places raises the frequency.
By changing the vocal tract away from a perfect tube, a singer can change the frequencies that it prefers to vibrate or resonate at. That is, by moving around the tongue body and the lips or by lowering the jaw, as in the formation of different vowels, a singer can change the position of the formants. The process of articulation determines the frequencies of the vocal formants, as every vowel is characterized and identified by its own unique formant frequencies.
The shape of the vocal tract is altered by the movements of the articulators (the tongue, jaw, lips and other facial muscles, mouth, and pharynx) in order to produce different sounds, such as those that occur in the formation of intelligible speech. These adjustments change the acoustic properties of the vocal tract. In other words, by changing the shape of the resonator, the quality of the sound is altered.
The formant frequencies depend on vocal tract length and shape, which are controlled by the positioning of the lips, jaw, larynx, velum (soft palate), tongue and pharyngeal side walls; that is, by articulation. There is a close relationship between vocal tract shape and formant constellation. For example, a shortening of the tract, caused by retracting the corners of the mouth or by raising the larynx, increases all formant frequencies, more or less. Both a narrowing of the pharynx and a widening of the mouth cavity increases the frequency of the first formant. This will make the sound brighter. Conversely, expanding the pharynx, protruding the lips and lowering the larynx will lower the first two formants and make the sound darker in timbre. Moreover, if the pharynx is lengthened by lowering the larynx, the second formant is also lowered in vowels produced by a forward position of the tongue, as in [i] (as in ‘heed’). However, vocal tract length, and hence the formant frequencies of a given vowel, varies somewhat both among and within men, women and children.
This method for adjusting the formant frequencies by modifying the shape of the vocal tract - the jaw, the tongue, the lip opening, the larynx, and the sidewalls of the pharynx - is commonly referred to as formant tuning (or formant technique or resonance tuning). It essentially entails raising and lowering formants in relation to the fundamental frequency in order to emphasize or de-emphasize one or the other, and thus create a more pleasant, balanced, vibrant singing tone. Formant tuning might be likened to the manual tuning of a radio dial, in which a listener slowly turns the dial until the sounds of the radio station become as clear as possible, free of static or other unpleasant noise. (In the case of singing, ‘noise’ represents any overtones in the voice that are non-integers of the fundamental frequency.)
Trained singers are able to control their vocal fold mechanism for accurate pitch specification, as well as the shape of their vocal tracts in order to modify formant structures for better resonance and projection. They are able to adjust their vowel formants to match one or more harmonics of the sung pitch. Experienced singers have thus learned how to ‘tune’ their formants over a reasonable range in order to make them coincide with the fundamental frequency or one of the overtones of the note that is being sung.
In addition to economizing the work of the vocal folds, the formant tuning technique allows a singer to extend his or her dynamic range, where simply singing higher or with more volume is limiting. One way to improve the tone and increase the volume (carrying power) of singing is to find an ideal and ‘comfortable’ vocal tract configuration empirically (i.e., the specific articulation that causes the biggest possible radiated intensity at each note of the scale).
Sundberg has identified portions of the vocal anatomy that he associates with the formant frequencies. The frequency of the first formant, which is responsible for ‘depth’ in the singing voice, is particularly sensitive to the size and shape of the pharynx and mouth cavity, which is largely influenced by the width of the jaw opening. The jaw opening constricts the vocal tract toward the glottal end and expands it toward the lip end. As the jaw is opened wider, this formant frequency is raised. Put simply, the first formant is responsive to the jaw opening. The first formant will usually contribute more to timbre because of its greater amplitude and lower frequency, closer to the fundamental.
The second formant is particularly sensitive to the shape of the body of the tongue. The second formant adds the vowel-defining element to speech.
The third formant is responsive to the position of the tip of the tongue and to the size of the cavity between the lower teeth and the tongue.
The fourth and fifth formants are more difficult to control by means of conscious adjustment of the vocal tract.
In the case of the Singer’s Formant – the cluster of formants found in some ranges of the singing voice in close proximity to the third formant - trained singers manipulate the third and fourth formants by lowering the larynx and elevating the tongue blade in order to enhance this part of the spectrum and make the voice able to be heard above an orchestral accompaniment. (These two formants can be made stronger in singing than in speaking.)
The formant with the lowest frequency is called f1, the second f2, and the third f3, etcetera.
The vocal tract resonances (formants) do not rise and fall with the phonated pitch produced by the vocal folds. These vocal tract resonances can only move up and down in pitch if the vocal tract itself is altered which, in turn, changes the resonance frequencies of the vocal tract. For example, sopranos frequently find it necessary to raise the level of the first formant (f1) because the fundamental pitch of a soprano often exceeds the normal frequency of the first formant. (The frequency range of the higher register of sopranos tends to overlap with the ranges of the two lowest formants.) This can create a shrill or thin sound in the upper register.
In order to raise the first formant, the vocal tract must be either shortened, or opened, or both. Female singers instinctively open their mouths as they sing pitches above the staff and, at the highest notes, they tend to raise the larynx slightly. All of these adjustments help to raise the first formant. Provided the fundamental frequency exceeds the standard frequency of the first formant (or even the second one), the formant tends to be dragged up to the fundamental.
While most sopranos may be able to vary the formants to follow the positions of the harmonics, the way in which this is done may vary between singers. For instance, some singers might raise the fourth formant in order to make it coincide with a harmonic, thus separating it from the lower formants, which typically might drop. Such a separation of third and fourth formants would prevent the development of a Singer's Formant. If the fourth formant drops along with the third formant, thus maintaining a close distance, the formants are reinforced. The presence of a Singer's Formant will not necessarily ensure high energy in that region of the spectrum; the spectral drop-off of the harmonics must also be sufficiently gradual.
As the soprano moves up the scale, her tone becomes increasingly instrument-like. The goal is to cultivate as ‘full’ a tone as she can, both in resonance and in registration.
The female voice is able to phonate a sound at the pitch level of G5 – the G above the staff - and higher, which is above the first formant frequency for the [i] vowel. If the fundamental frequency of the underlying vibration is higher than the formant frequency of the system, then the character of the sound imparted by the formant frequencies will be mostly lost. There is a marked tendency in women’s voices for the high formant to drop out at some point as the pitch rises. Coloratura voices sometimes have practically no high formant at all.
The higher a soprano sings, the more her resonators simply reinforce her fundamental and the less they function as vowel formants. If the pitch of the phonated sound is above the first formant (the ‘fingerprint’) of the selected vowel, the first formant of that vowel will not be present, and the vowel will not be easily identified. This is most apparent in the example of soprano opera singers, who sing high enough that their vowels become very hard to distinguish, which explains why female opera singing is often difficult to understand, as a consequence of the harmonics and formant patterns. This is why coloratura arias usually abandon all words and simply turn into vocalizes (wordless vocal ‘gymnastics’).
In untrained singers, the highest notes of the female voice often show distortion - a preponderance of upper partials, which creates a thin, shrill sound - whereas in premier female artists, the oscuro (‘dark’) aspect of the chiaroscuro tone never diminishes. In fact, when these prominent singers are singing in upper range, the first formant and the fundamental are often enhanced and exhibit increased acoustic energy in the lower portion of the spectrum. This balanced tone is achieved because classically trained sopranos make use of the technique of resonance (or formant) tuning to intensify the vibrations of the vocal folds, to increase the power of the voice and to balance tone.
Opening the vocal tract will raise the first formant, but raising the first formant will place it in the pitch area of the first formant for another vowel, [a]. Consequently, the most accessible solution is to use the first formant frequency of the [a] vowel, which is the highest first formant for any of the vowels, and, in addition, attempt to raise that formant of that vowel by opening the mouth a little wider. Sopranos do this instinctively and correctly. However, the vowel is no longer [i]. It has now become some form of the [a] vowel. (This technique is also part of vowel modification, which I will write more about in the continuation of this article.)
The soprano can open her mouth wider and still keep the back of the tongue high in a quasi [i] vowel position, and this will continue to give a bit of the [i] quality to the notes sung above the G5. She will be able, through this adjusted position, to resonate the f2 formant of the [i] vowel, but nothing that she does will make it possible for her resonate the f1 formant of the [i] vowel because the f1 formant is lower than the pitch that she is phonating. Because the f1 formant is the strongest and most decisive formant in defining any vowel, resonating only the f2 formant can create only a suggestion of a vowel.
However, integrity of the vowel must be preserved. The primary mouth shape for each vowel survives, but the mandible (jaw) gradually drops just enough to avoid the conjoining of high pitch and lateral vowel as pitch elevates. Extent of buccal aperture (mouth opening) in high range must match both the phonetic demands of each vowel and the pitch being sung. Increased dynamic intensity (including volume) also commonly contributes to an additional degree of mouth opening.
Vowel definition must always be recognizable, yet vowel modification remains quintessential to upper-range singing. If the singer fails to adjust her mouth and vocal tract to permit vowel intelligibility – that is, if she attempts to maintain the same lateral mouth posture that is appropriate to the lower registers, holding her mouth in an unmodified lateral position while singing in the upper range – the first formant will be forfeited and her resonance balance will be upset. She will lose the ‘velvet’, the ‘richness’, of the voice and she will produce an overly bright, shrill or thin timbre and destroy diction, (not to mention experience difficulties fully accessing the head register because of tension).
Lowering the jaw while forming a front vowel enables the singer to approach a back vowel position, which lowers the intensity of the upper harmonic partials. (Lowering the jaw lowers all formants; the lateral front vowels [i] and [e] enhance acoustic strength in the upper portion of the spectrum.) If the apex of the tongue remains in the lateral vowel postures while the jaw lowers, the singer will have an appropriate vowel modification. (The phonetic shape of the vowels changes somewhat.)
However, if too large a buccal opening is assumed, disequilibrium among overtones will also be produced. (I have written more about ideal mouth shaping, above.)
Formants are vocal tract resonances that appear at certain frequencies - the formant frequencies – and are measurable peaks of acoustic strength. The energy in a formant comes from the sound source (e.g., the periodic vibration of the vocal folds). Within each formant, and typically across all active formants, there is a characteristic waxing and waning of energy in all frequencies that is caused by the repetitive opening and closing of the vocal folds.
The vocal tract filters a source sound (e.g., periodic voice vibrations or aperiodic hissing) and the result of the filtering is a sound that can be heard and recorded outside the lips and shown on a spectrogram. Each of the preferred resonating frequencies of the vocal tract – each formant - creates a ‘bump’ in the frequency response curve. Formants occur, and are seen on spectrograms, around frequencies that correspond to the resonances of the vocal tract.
During formant tracking, analysis of the singing voice is provided through electronic devices (such as spectrograms) that measure peaks in the harmonic spectrum of the voice during singing; that is, formants are measured and analyzed. Singers use these electric signals to get a better idea of the quality of sound that they are producing – whether or not certain formants are present, whether certain formants are stronger than others, etc. - and then make adjustments to how they sing through modifications of the vocal tract in order improve the acoustical quality of their voices. (This technique is known as formant tuning.)
Through the process of formant tracking, singers and their teachers can interpret signals and connect them with the traditional goals of voice building, such as improving vocal quality and carrying power, equalizing the vowels, optimizing vibrato and legato, extending the useful range and the joining the registers.
VoceVista-Pro is a software system, with optional hardware, that is designed primarily for the analysis of the singing voice and providing feedback during singing instruction. It utilizes two non-invasive signals – a microphone and a miniature EGG (electroglottograph) – allowing singers to examine their voice production in real time wherever they have access to a laptop computer.
(Disclaimer: My mention of the above software system should not be misconstrued as a personal endorsement of that particular product, as I have neither fully researched nor practically tested it, and therefore cannot vouch for its usefulness or effectiveness. My purpose for mentioning the name of the specific product is merely to inform my readers of the existence and availability of such tools for singing. It is up to individual singers to do their own research before purchasing such products to use in their vocal training.)