In the Opera world, one of the most exciting things to anticipate and hear is the brilliant, climactic high note of the tenor soloist in an aria. Not only does the voice carry well without amplification, but takes on a distinctly thrilling, impressive quality of resonance that other parts of the voice do not quite have in the same way. In this post, I will explore the ways that these fine singers manage their voices to enable such singing.
Introduction
Since we will be objectively discussing vocal tone quality, I will be using spectrographs to assist. With some of these particular ones, I will include annotations to the images so that the reader can make the connection between the visual representation and aural experience of harmonics within the vocal tone quality. The spectrographs I use will all be of the final note in the tenor aria, 'Celeste Aida', from Verdi's Opera Aida, which is on the syllable 'Sol' on the Bb above middle C. To give credit where credit is due, my investigation in this area was inspired by the published work of Donald Miller at www.vocevista.com. The spectrographs were produced with Spectrogram16, by Richard Horne.
Bjoerling and Domingo
The spectragraph to the left shows that note from recordings of two of the most popular and capable operatic tenors of the 20th Century, Jussi Bjoerling (represented with the blue line) and Placido Domingo (represented with the white line.)
To help orient you to the image, I have annotated it with lines and text to show the locations of the harmonics of the sung tones. On this diagram, left = lower frequency, right = higher frequency. Up = higher intensity, down = lower intensity. The frequency range represented is 0 to 4000 cycles per second (Hz). So that these notes could be compared as well as can be from recordings, I equalized the volume of the fundamentals.
What We Can See and Conclude
With this equalization, the fundamentals and 2nd harmonics (H2) are about the same strength when comparing voice-to-voice, as evidenced by the nearly exact overlay of the blue and white lines. However, a very great difference is noticeable in the intensity of H3. Bjoerling's H3 goes way higher on the intensity scale than Domingo's, indicating that it is very much stronger. H4 and H5 are also more intense than those of Domingo, though their intensity in Domingo's voice increases until they are in rough parity with that of Bjoerling at H6. From there, the intensity of harmonics falls off dramatically in both voices.
So, as a proportion of the overall sound of the recorded voice, Jussi Bjoerling's tone quality and power are created mostly by H3, H6, H5 and H2-- in decreasing order by intensity -- while Domingo's tone quality and power are created mostly via harmonics H6, H2, H1 and H5: again in decreasing order by intensity. These different balances, while they both sound like tenors, make them distinguishable to our ears.
What We Cannot Conclude
Does this mean that Bjoerling's voice was "bigger" or more resonant than Domingo's, or perhaps the other way around? Neither one! Engineers who make recordings adjust volumes and balances at their own discretion, to make recordings have a satisfying overall effect for the listener, while not overwhelming the recording or playback machines. There is simply no way to tell from a recording what the original sound intensities were, only how they were after they were recorded and mixed down. Sometimes, though not much with Opera, some EQ is added to overcome a recording problem, or to sweeten the effect a bit. Some of that latter can be seen in some of the images here, and is discussed below under the section "Engineering Artifacts".
So, even though we cannot learn the size of these voices in absolute terms, we can learn (in general) how the sound energy of the harmonics is distributed relative to one another within a single recorded voice, and can compare recording to recording.
Vocal Resonance Strategies
Vocal power that is distributed across the various harmonics is perceived by the listener differently, according to the frequency range of the particular harmonics. In the case of the Bjoerling and Domingo notes, the reason that there is such a dispartity in the displays of the blue and white lines is that these singers have balanced their resonances differently for this note in the recordings selected.
Surveying recordings of more than 40 of the top tenors of the 20th Century, these voices predominately use one or both of two strategies to create the powerful top voice. In this next section, we will explore the strategies that they used, and comment on the overall effect.
The "Singers Formant" Region
Looking back at the picture for a moment, you may notice the two vertical red lines which bracket the frequency range of the 6th Harmonic, very strong in both voices. These lines show the center 400Hz of the singer's formant region, and also indicate the area of highest hearing sensitivity. When harmonics are strong in this frequency region, they are very audible, adding to the carrying power of the voice and to the listener's perception of voice quality as well. For the singer without amplification, presence of these frequencies allows the voice to cut through above the sound of a piano easily, and even a full orchestra in the concert or Operatic venues. These frequencies also help the audience member locate the sound source very specifically on stage, a big help when singing an ensemble. Both Domingo and Bjoerling have this important feature in their voices. Incidentally, the frequency of the 6th harmonic is two octaves and a major third above the sung fundamental.
The first most common strategy for vocal power and audibility is to have a strong singer's formant, as strong or stronger than the fundamental and 2nd harmonic. We could also call this the 'high ring' strategy.
Lowest 3 Harmonics
The perception of the "darkness" or "warmth" of the voice comes from the intensities of the lowest two harmonics, H1 and H2, which are the fundamental of the sung tone, and the octave above it. For these, both singers have about the same proportion, and this forms a solid core to the sound in both voices. To the listener, these two harmonics are very difficult to distinguish individually when they are approximately the same volume.
The presence of the proportionally louder H3 in Bjoerling's voice introduces an interesting difference. H3 is the frequency an octave and a perfect 5th above the fundamental, what (to a classical organist) would be called a 'quint'. This quite strong harmonic colors the tone distinctively, and, because it is an odd-numbered harmonic, it stands out in the awareness of the listener, adding brilliance to the vowel. When the 3rd harmonic is the loudest in the whole voice (such as it is for Bjoerling) this becomes a significant feature of the tone quality, and carries a great deal of the vocal power.
The second most common strategy for vocal power and coloring is to have a strong 3rd Harmonic. The strong H3 is obtained by singing a vowel which tunes the 2nd formant (F2) to just a little bit higher than H3, a process sometimes called vowel modification, or vocal tract tuning. We could also call this the mid+high ring strategy. (Note: For other combinations of note and vowel impression, the tuning of F2 is more advantageously made to H4.)
In professional voices, both of these individual strategies can be found, and also combined. Jussi Bjoerling is a fine example of the combined, and Placido Domingo is an excellent example of the "singer's formant" or low+high ring strategy.
Another Singer for Comparison - Franco Corelli
Franco Corelli is known for a heroic tenor voice. This spectrogram shows the relative strength of the harmonics in his voice for the same note we were examining with Domingo and Bjoerling. Though there is a bit more orchestral clutter in the sample -- sharp spikes here and there -- and on the left end) you can see clearly that the 3rd harmonic is very prominent in his voice, . Looking to the right, you see some strength with H4, H5 and H6, and then a strong H7 as well. This would make his approach a combined one.
Here are some other spectrograms for comparison. See if you can identify which strategies they employ:
Alfredo Kraus
Benjamino Gigli
Luciano Pavarotti
Special note here: Pavarotti's voice is very interesting in that he uses the H3 formant tuning, but does not combine it with a strong singer's formant. The overall effect is very distinctive.
Enrico Caruso
Mario Lanza
Engineering Artifacts - Possible
The clustering of the formants F3, F3 and F5, which combine in the singer's formant region ordinarily produce somewhat jagged peaks in a spectrographic display. When recorded and displayed as is, without any sweetening EQ, they do not often take the shape of smooth curves, rounded on top, but will ramp up and down fairly sharply across three or four harmonics.
Go back to the Kraus spectrograph, and look at the shape of the curve created by the tops of H4, H5, H6, H7 and H8. Disregard the leading (rightmost edge) pointy peaks that show up, that is an orchestral note. The "wide" part is from the voice. In my opinion, the slow ramp-up of the harmonic intensities in this region, peaking at H7, and then diminishing a bit to H8, just looks too regular. I think this is a likely example of some EQ shaping to allow the voice to cut through the orchestral mix.
Though I cannot be quite so sure on this one, the suddenly very strong H7 in the Correlli spectrograph looks a bit out of place, with the intensities of the immediately three lower harmonics at the levels they are. Now you know what you might look for, I will leave the judgment to you. It's not likely, while listening to the recording, that you would be aware of any of these harmonics individually, anyway.
None of these latter points reflects on the quality of the singer in any way, nor would the singer likely have been aware that tweaks were done on their behalf. As I said earlier, the Engineers work to create an effective recording of the voice that fairly represents what the performance sounded like to them.
Summary
We've seen with these examples the most often occurring resonance strategies for creating the ringing top notes of the Operatic Tenor voice, and readily accessible examples from some of the most popular singers of the 20th Century. We've also discussed the limitations of using recordings to make these conclusions. If you'd like to see more articles of this type, studying the vocalism of other voices, please send me a comment as to your interests. In any case, I plan to do a parallel discussion of the resonance strategies of the Operatic Baritone (Warren, Milnes, Tibett and Bastianini!), the female high voice, and discuss in detail the challenges involved with the transition from mid-voice to the top in both types.
A Christmas Egg
The following spectrogram is of Michael Bolton (in blue) and Luciano Pavarotti (in white) singing the climactic note of Puccini's aria 'Nessun Dorma' from Turandot. These were public, large-hall performances, and the performers were close-miked, a very interesting way to hear Pavarotti's voice. The note being sung is the B natural above middle C.
A problem I encountered in comparing the voices with these recordings is that orchestra is playing quite loudly, so the first harmonics are cluttered by those sounds, so much so that we cannot really distinguish what component of the sound is the singer, and what is the orchestra. To do this particular equalization, I matched volume of harmonics H2 (right above the 1 on the bottom scale) and H3 (midway between 1 and 2), since the vocal vibrato in both voices makes the trace wide enough to see.
Interesting, that Michael Bolton and Luciano Pavarotti have almost exactly the same resonance balance ratio for these two harmonics. Remember, this sort of comparison does not tell us about the absolute volume of the voices, just how the resonances are balanced. You can see some places in the higher harmonics where Bolton's voice has relative strength, too. He has characteristic singer's formant strength that peaks at H6, (right in the sweet spot of our hearing) which would make his resonance strategy for the note a "combined" one, from our former terminology.
If you are interested to listen to these performances, they are at
Michael Bolton to about 3:15 into the song and Luciano Pavarotti.
This essay was first published December 12, 2008 on The Modern Vocalist.com the Internet's #1 community for vocal professionals, voice health practitioners and pro-audio companies worldwide since November 2008.
Report Record
