Wednesday, February 4, 2009

The female middle voice, a complex and misunderstood puzzle, Part 2

This portion of the discussion is surprisingly problematic. I will deal with the range F4 to F5 inclusive. This is called the middle register because it lies in the middle of the total female range. The fully-functional female range has a modal span beginning at F3 (a fifth below middle C)and ending at F6 (the Queen of the Night's famous high F). This is controversial nowadays because not many singers access those notes. That is because the flute voice is left undeveloped and therefore a correct muscular balance in the extreme high range becomes difficult. I will deal with the female high range in a future post, but suffices it to say that the upper end is rarely fully explored.

As for the middle range, as always, we must consider both the muscular and acoustic problems.

1. Muscular: We have already established that the "real voice" (the modal dynamic that involves muscular antagonism between vocalis and CT) spans about 3 octaves. This means that halfway through that range should be the muscular shift. I have been of the mind set that the muscular shift was around F5, but this does not compute. The second passaggio (between E5-G5) is an acoustic shift where the first formant takes over the resonance again. The muscular shift probably happens around C5. A colleague of mine in my early days of teaching in Utah mentioned this once and I did not comment on it. It seemed logical then, but I did not have the information to back it up. But when we think of it, it is around C5 or there about that female voices reach a lighter mechanism. If a woman takes what she calls the chest voice up, the voice usually will crack around Bb4-C5 where the muscular shift should occur. That is why Agnes Baltsa exhibited a muscular shift around C5 in the Carmen clip featured in the previous post.

If she had gradually lightened up (and this includes managing the pressure effects of the first acoustic passaggio), she would have accomplished a smooth climb through the C5 area get through the rest of the voice (having to deal with the mild acoustic issue of the second passaggio).

Maria Ewing:

Ewing uses a very dynamic approach to the muscular shift. The first verse shows a careful expert muscular dynamic, almost textbook. As the song becomes more crazed she makes use of a heavier approach, like Baltsa, unapologetically violating the muscular shift.

Marina Domashenko:

This is a more traditional vocal approach to this vocal dare. The Russian/Slavic mezzos tend to be heavier and have natural weight in the low range and do not feel a need to add more weight for color.

Ewing was a lyric mezzo who eventually tried her voice less successfully at Soprano repertoire. This speaks to the nature of her voice (i.e. lighter mezzo). Lighter mezzos have a tendency of overdoing the low range to compensate for the lack of power in that range in order to produce sounds they feel are necessary to the role in question. Carmen poses such problems.

Both Ewing and Baltsa being such extraordinary actresses are able to make an art of the register violations. One could argue that the register violations support Carmen's rebellious anti-establishment nature. No counter-argument from me from an acting standpoint! The problem from a pedagogical standpoint is that routine register violations in a certain role makes more traditional roles that require vocal balance very difficult to sing.

Grace Bumbry:

Grace Bumbry, a dynamic singing actress in her own right gives a very balance reading here. The reckless third verse is also chestier but her natural vocal weight is heavier than Ewing and Baltsa, so the register violation is minimized.

Elina Garanca:

Of the lyric mezzos around today, I am a fan of Elina Garanca for not resorting to vocal compromise to get her dramatic effects. The voice gains in power over the years precisely because of her technical discipline. This version of the "Gypsy Song" is dramatically committed but does not violate her vocal balance. She reserves the chestier colors for the lowest parts of the range where it is appropriate and achieves dynamic (i.e. gradual) balance through the middle range as opposed to static (i.e. sudden) changes.

But today's audiences are less knowledgeable about the difficult muscular balance required to sing a strong and safe middle range. Even opera singer themselves are more excited by the dramatic though unhealthy effects of habitual register violation that produces extreme colors in the lower middle at the expense of balance and power in the A4-C4 area.

2. Acoustic (Resonance). While analyzing many voices over the years I discovered surprisingly that female classical singers do not conform to the acoustic norms that we expect in the middle range.

Ideally, F3 to E4 should be first formant dominant and it is in practically all female voices. F5 to F6 should be mostly first formant dominant and it is. F4 to E5 should be second formant dominant and in general it is NOT. And that is one of the reasons that the middle range is chronically weak sounding among female classical singers.

The one thing that separates the classical singer to the good pop singer is resonance strategy. Operatic singing requires a sound balanced between low and high overtones. This "chiaroscuro" balance requires a released (i.e. lower) laryngeal depth. By maintaining this depth (all other functions being correct) the resonance mode should change spontaneously from first formant dominance to second formant dominance. But this rarely occurs because we are used to first formant resonance and it gives the voice a sensation closer to speech. And this aspect is an acoustic determination of what we deem "popular" and not.

Opera singers have always had a desire to reach a broader audience. Tenors singing Neapolitan songs, African-American singers singing Negro Spirituals, German singers singing Operetta hits, etc. Pavarotti who defined a "real tenor" based on his ability to "cover" (accessing second formant resonance at the passaggio point)violated his rule often when he sang popular Italian songs.

Here is Pavarotti's "cover manifesto" at 3:15:

Then his register violation (in classical terms) in the beautiful song Caruso:

The acoustic problem with the female middle voice is that the adjustments necessary for the ideal second formant tuning is not so easily achieved. In fact most female singers go in and out of second formant tuning in the middle range. Because of this difficulty many singers attempt to get closer to the first formant tuning (belt tuning) or more commonly some non-resonant tuning between the two. This is why the female middle voice is so often not resonant. The best that can be found among our most accomplished operatic divas is a high percentage of second formant tuning in the middle range. Even the best singers tend to revert to 1st formant or non-resonant adjustments.

In short the best middle voice models are rarities both in terms of muscular adjustments and in terms of acoustic adjustments. When I was preparing this post, I expected to find many singers on that I could acoustically analyze to prove my point. I was disappointed. The next question then could be that my expectations are wrong. I thought of this. But acoustic theory is based on pitch level and its interaction with the vocal tract (vowels). It is a relatively simple thing to observe relative to the limits of vowel formants. Formant dominance in the male voice tends to be easier to distinguish. Male singers either correctly tune to F2 above their respective passaggio or else remain F1-dominant. They do not tend to go back and forth. The reason for this can only be one thing, namely that male voices sing in a range conducive to vocalic intelligibility. Even the F2 range for male voices requires only mild modification. Male singers also accept that high notes necessitate modification and that the viability of the resonance adjustment for high notes trumps vocalic purity. Women will conform to this in their high range but find it more difficult to accept that substantial modification of vowels is necessary in their middle range. Many women do not believe that such extremely modified vowels will sound intelligible in context. So they sabotage their own resonance viability for a feeling of vocalic satisfaction. Paradoxically, that feeling of vocalic satisfaction is precisely what hinders viable resonance and therefore any real chance at intelligibility in context.

© 02/03/2009


Martin Berggren said...

Thanks for an, as usual, interesting and intriguing post!

Something I have asked myself is what Pavarotti's demonstration of "white" vs. "covered" sound really demonstrates! Does it demonstrate the difference between 1st and 2nd formant dominance? This is what I thought when I first heard it. However, I was really surprised to find out when analyzing the sound that I was wrong!

When Pavarotti demonstrates a "white" sounds, it is H4 that completely dominates the spectrum. In the covered version, H4 is very weak and it is instead H2 and H3, at about at the same level, that dominates the spectrum. For first-formant dominant singers (e.g. Bergonzi), it is H2 that dominates, and for such singers, the spectrum falls of very quickly, leaving very little energy above H2 and up to the region of the singer's formant.

If I am allowed to speculate, I believe that a successful acoustical turning of the voice in the passaggio (something I am much too slowly learning to do myself!) requires the basic skill of producing strong lower harmonics, H2, H3 & H4 (well, there's the issue of the singer's formant as well, which I believe is connected somehow). Then, vowel modifications can be applied in order to balance the levels of these harmonics, in order to obtain a "noble" tenor sound or a more "white" pop-like sound.


Toreadorssong's Vocal Technique Blog said...

Thank you Martin for another excellent observation.

Quite often a singer makes an adjustment that feels right for him and has no idea what to call it or how his wording is understood. I believe that clip of Pavarotti speaking about covering is almost contradictory as the sound he calls too open or spread (white) ends up being F2-dominant anyway (from your description of H4 dominating the spectrum).

What he calls covering particularly around the passaggio requires a balance between the two formants, and that is the more important thing. This is also true of the female middle voice. It is not enough to tract F1 or F2 it is important that there be a balanced relationship between them. I firmly agree with you. Even when F2 takes over in the male high range, it must never be so high above F1. I find this creates "dirty" peaks between the harmonics. Balance between low and high partials is crucial. When there is balance between F1 and F2 I find greater strength in the SF.

Thanks for bringing this up.

Martin Berggren said...

Well, what I wanted to suggest - and something I wish to get your opinion about - is that the successful acoustical "turning" of the voice in the passaggio may not be a question of formant tuning alone. I suspect that the phonation quality is even more fundamental, that is, how the vocal folds and the region just above the larynx is postured. I noticed when recording my own voice that the spectral strength fell off quite sharply above 1kHz and picked up again around the singer's formant region (2.5-3 kHz). This kind of spectrum allows the first formant to boost H2 for most of the range, so everything may feel all right, and there is a certain ping about the voice all the way up through the passaggio. However, there is not enough higher harmonics for the second formant to naturally take over at the passaggio. In order to get a ringing top with great response of H3 or H4, I believe there has to be an adequate generation of a broader set of harmonics all over the vocal range. That means that in the middle range, the H2, first-formant dominance in the middle voice should not be too overwhelming, and that there should a certain presence of higher harmonics also in that range, making it easier to shift to second-formant dominance above the passaggio.

Does this make sense, or am I at the wrong track?


Toreadorssong's Vocal Technique Blog said...

I agree 100%. It is completely logical what you are suggesting. The fold posture is at the root of everything I have discussed here lately. I believe there is an optimal length of the folds for a given F0. Too long (thin) or too short (thick) will result in a muscular dynamic that postures the folds inefficiently for the given F0. I believe the Vocalis-CT balance is the primary muscular concern in the larynx. This balance influences how the rest of the muscles behave (providing the singer has the right sound in mind. Everything results from the concept of the sound). When the balance is correct there is a sensation of both anchoring and release, a sensation of firmness and flexibility simultaneously. The fold posture influences air pressure/flow balance as well as the facility of accomplishing favorable acoustic adjustments. The sensation of transition from chest voice to head voice must be one that is very gradual. Sudden changes are not symptomatic of balance.

All the best!

Adam said...

I found this post very interesting, and I have a question for you. Now, I am not as informed as you or as the above commenter is about the acoustics of all this, though I can fudge my way through a discussion of formants and harmonics well enough to get the gist.

You said that men accept that their sound must be modified to make it into the vocal coordination above modal/chest/heavy/whatever you want to call it. Thus, they either remain F1 dominant, or switch to F2 dominant sounds, in that range.

As a bass-baritone-ish voice (I'm not a huge fan of labelling voices and putting them in neatly arranged categories), I'm having trouble around the region of D4-G4. That is when I shift from a TA-dominant mechanism to a CT-dominant mechanism, and I don't have severe problems with making the handoff of muscle activation smooth, as long as I'm not trying to sing loudly. It's something for me to work on, but everyone works on that. Here's the problem I'm dealing with: the sound becomes very weak and somewhat breathy for about a third above the middle of my transition area. I've thought that I wasn't getting proper cord adduction and I've done my adduction exercises, but they don't help here. In fact, I no longer think adduction is a problem, because I have a powerful sound in the upper reaches of that mechanism. I wanted to know what you thought of this, relative to the discussion of the transition from an F1 dominant to an F2 dominant sound. Do you think that the problem I experience in this range is that I'm not making any of the formants strong enough, and I'm left with a nonresonant sound, or is it some laryngeal (as in, not supraglottic) maladjustment around the transition area that results in a loss of sound output?

I much appreciate your help,

Toreadorssong's Vocal Technique Blog said...

Dear Adam,

Thank you for a terrific, well-expressed post. Without hearing you it is hard to make conclusions, however your problem sounds like what I encounter with many Bass-baritone (ish) types. You will tend to have power in the upper voice rather you sing it correctly or not, but that too is relative. The folds are brought closer together when they are stretched. The external TAs (not the vocalis which are internal) plays a part in fold adduction as the pitch rises. That is one of the reasons we are able to have full closure falsetto in the extreme high range. Other function like vocalis activity are not indispensible there (although better for a balanced sound).

The thing that comes to mind is the nature of fold posture relative to fold closure. I recommend singing [u-i] on A3 with a light head voice (not breathy not pressed) then crescendo and diminuendo. The [u] will induce laryngeal depth and the [i] will induce fold closure without excessive pressure. I believe this combination sets the resonance of the vocal tract in such a way that induces inertial reactance whereby the air on top of the folds help with closure. Once you experience this balance, then attempt to maintain both the open feeling of the [u] and the pointed feeling of the [i] when singing [a]. Maintain this sensation in the upper baritone range is difficult but can be achieved. Finish the exercise by doing [u-i-a]. Let me know how it goes. Remember do not force the onset. Smooth, falsetto-like but clear and not breathy.

Michel Galán said...

hey men, how you do then the passaggio you should cover? the C5 or the C4 or both?

Kashu-Do said...

Michel, are you talking about the female voice? Covering is a loaded word! C4 is not a problem. C5 may require vowel modification but it is more an issue of muscular balance and breath flow!

Math Flair said...

Hi Jean-Ronald, I have an important question about the female passagio:

We know there is a muscle shift around C5, where the second formant becomes dominant. But I guess this shift actually happens gradually in the whole F4 to E5 passaggio, right?

My question is: is there another muscular shift starting at around E5 or F5?
I do feel that the coordination of "CT muscle dominance and ligament vibration" changes there and wants to work with something else, whilst shifting back into the the First Formant.

(From Quotes: " This means that halfway through that range should be the muscular shift."

"Ideally, F3 to E4 should be first formant dominant and it is in practically all female voices. F5 to F6 should be mostly first formant dominant and it is. F4 to E5 should be second formant dominant and in general it is NOT. And that is one of the reasons that the middle range is chronically weak sounding among female classical singers.")

Thank you!