In Search of the Soprano Singer's Formant: A Perceptual Approach

This video is of Dr. Ian Howell's presentation at the 2017 Society for Music Perception and Cognition Conference in San Diego. This material is a combination of his model for understanding the special psychoacoustics of the singing voice, as found in his VOICEPrints article, "Necessary roughness in the voice pedagogy classroom: The special psychoacoustics of the singing voice," (www.academia.e..., and research in collaboration with Rebecca Worthington , GD, MM.
www.necmusic.ed...
The spectrograph used in this video is Overtone Analyzer. The company that makes it has just recently merged with VoceVista. If you are interested in exploring sound with this program, the equivalent program available from the new company is called VoceVista Video Pro. More info can be found here: www.sygyt.com
Abstract:
One critical difference between speech and classical singing is the presence of the "Singer's Formant" (Fs). Bartholomew (1935), Vennard (1967), Sundberg (1987), McCoy (2005), Bozeman (2013), and others locate the Fs phenomenon below approximately 4kHz, placing it within the frequency band preferred for speech comprehension. Sundberg demonstrates that the third, fourth, and fifth vocal tract resonances cluster to profoundly amplify nearby voice source harmonics. At pitches typically sung by classical male, non treble singers, multiple harmonics fall within this boosted frequency range. Since the number of harmonics present within this frequency range necessarily decreases as pitch rises, and harmonics must be present to express vocal tract resonances, the voice pedagogy literature tends to present the Fs as a characteristic of the male singing voice, absent in the female voice above C5.
If, however, the Fs is defined neither by frequency range nor specific vocal tract resonances, but by perceptual qualities, one may begin to search for harmonics exhibiting similar qualities above C5 in the elite soprano voice. Auditory roughness, a buzzy quality present when harmonics fall within critical bands, is generally related to position within the harmonic series rather than frequency range. A tenor singing Bb4 may have as many as four harmonics in a critical band below 4kHz (his observed Fs). A soprano's F5 has no roughness in sub-4kHz harmonics. However, a spectrogram of an elite female singing above the treble staff frequently reveals high frequency spectral peaks in the range higher than typically studied in speech perception. They are composed of harmonics that fall within critical bands, a key perceptual quality of the male Fs.
This presentation features analysis from a pilot study (n=4) of advanced graduate level sopranos demonstrating Fs-like peaks that exhibit acoustic roughness in the elite female voice, albeit at higher frequencies than in males. Defining the Fs perceptually and searching for such qualities where they arise in the female voice---analysis of what is present rather than what is missing vis a vis a male voice---allows one to begin to consider how these Fs-like peaks are characteristic of, rather than incidental to the timbre of the elite female singing voice.

[@rarevisionog]

Judging it by the ear seems to be very subjective unless those with very sharp hearing. Many singers in pop they say to have the ring, may not show it on the spectrogram

[@bflkel]

this is awesome and interesting

[@bflkel]

I wonder if our neglect of the female "singer's formant" would have been as long as it was if they had simply made pianos go an octave higher to begin with, so that we were more knowingly familiar with that spectral range and could reproduce it on demand. In hindsight, based on my understanding of how *everyone's* overtones work, this should be obvious that there is contribution from up there no matter what, right?

[@Kurdyukov87pianist]

Type High singing formants and lowest notes: Soprano 3500 Hz G#3,A3 Mezzo soprano 3200 Hz F3, F#3 Contralto 3000 Hz C#3, D3 (and below) Tenor 2700 Hz G#2, A2 Baritone 2500 Hz F2, F#2 Bass 2300 Hz C#2, D2 (and below)

[@Kurdyukov87pianist]

Some contraltos sounds like sopranos if they have the second high singing formant from 3500 Hz and higher. For example: Whitney Houston, Mariah Carey...

[@fernandolazari2339]

Thanks Master!!! Can you send me information about the program?

[@benduru1278]

Hello, I have a question. Do you have a video on the tenor voice? I'm interested in learning about certain vocal exercises that take me into and out of head voice, developing a strong flexible head voice and certain exercises that work on mixing. Thank you so much.

[@FilipeGeyerVocalStudio]

So... nowadays I hear that the singer's formant is not important and it is not even desirable, but only a genetics thing, you either have it or you don't. What you should look for instead is people that have a higher 3rd (or 4th) harmonic compared to the second. Michael Jackson and Pavarotti did that. Domingos, who had a S.F., didn't do that. Cool video!

[@mradaChris]

What is that app you are using? I would love to use it myself.

[@singwisevocals]

Brilliant, Ian!

[@mradaChris]

I really enjoyed this. I'd like to know what the name of that software for the spectrogram is called. Mine, Spectrogram 16, is quite old and graphically clunky.

[@rentourapartment]

Chris Sparks hi Chris :-). This is a software called Overtone Analyzer. The company just merged with VoceVista, and the equivalent software is now called VoceVista video pro. I’ll put a link in the description.

[@mradaChris]

Thanks for the link! It works ok on my PC. Much crisper indicators for the frequencies. So I have a question, does a formant consistent of several frequencies as display on the graph?

[@singwisevocals]

Ian - Is this version of Voce Vista the same as the one that came with my copy of Donald Miller's book Resonance in Singing?

[@rentourapartment]

Chris Sparks That’s a really good question, and it’s a little bit confusing. We actually use that term to describe two different but related phenomena: The resonances of the vocal tract, and the peaks in the radiated sound. In voice pedagogy, we typically use the term formant to mean the first definition. However, the term singers formant cluster is interesting because it describes a single radiated peak that is actually caused by the clustering of three vocal tract resonances.

[@holophrazeinikos2046]

This is fascinating stuff. My question is (and I'm not at all sure there's a simple answer): Why are 'elite' classical singers, on the whole, less skilled (technically adept) in the periods before a great deal of scientific information was well known? Some may disagree with this premise-but many would agree that there are few singers of the Caruso, Pavarotti, Sutherland, Ponselle (and numerous others available on video,recording and youtube) among today's performers. These great singers were taught without reference to this information. One famous pedagog observed "Now they are finding out how the system works- but they no longer know how to work it"

[@rentourapartment]

Holophrazeinikos This is an often repeated question in the classical singing community, and while I’m not sure there will ever be a decisive answer, I personally reject the premise. In the version of the question the way that you framed it, we take the absolute best singers of previous generations, and hold them up as examples of what everyone from their time sounded like. I think it is easy to frame the question in terms of the impact of scientific inquiry, and that it has had a damaging effect on actual singing training, but I don’t think there’s evidence to bear that out. I think some people are good teachers connected to a tradition and others are not. And all those people understand the science of singing to varying degrees. I am fairly certain that at every moment of history, those considering the singing voice lamented the then present decline in light of some imagined golden age. I think it’s really just a sampling error. We don’t hear all the substandard singers from the past, because their sounds were never captured, nor their voices celebrated in print. The other thing to consider, is that the economic model of classical singing has changed dramatically over the past century. This includes the advent of commercial jet travel and the expectation of a packed schedule without downtime.

[@holophrazeinikos2046]

I appreciate your thoughtful and well-considered response. Thanks- I agree with much you say - When comparing the highest level(completely subjective) level of current performers with their superstandard counterparts in other periods, I personally hear a generally far lower competency in many parameters-not only quality of sound and expressivity, but things like pitch, rhythm, articulation, coloristic variety and others. Perhaps you disagree- but that's cool. Scientific info per se clearly can't hurt anything, but it's misuse, surely unintentional in most cases, has in many fields of endeavor. Certainly research can be a valuable tool. But the skill of the technician is a major factor. Your later comments about economic model and travel without downtime ring very true. Academic institutions-relatively recent forums for voice cultivation- must have altered the situation as well. I applaud your work. A lot of bad things have been done in the name of science That doesn't make it science's fault- Haha Thanks again for your response and Best Wishes in your continuing work.

[@rentourapartment]

Hi Stone, thx for this comment. I think that you illustrate a common perspective, which (if I have understood you) is that voice science is synonymous with a mechanistic and anatomical approach. There is a lot more to it than that, I think. Much of it touches on cognitive and motor pathway issues as you mention, and a lot of it deals with aspects of airflow and perception that are highly useful for singers. I can't refute the claim that singers of the past were superior or that life was simpler, but I also don't think there's evidence to support it beyond specific recordings (the sampling error) and our personal feelings. Again, because we do not have a comprehensive, culture-wide view of what singing was. Are there clear examples of people training voices poorly today? Absolutely. I think the burden of proof that the past had no bad voice teaching or bad singing falls on the one making that claim. And I have to think too of the negative impact to singers of the past simply due to advances in medicine and surgery. Perhaps we have terrible singing now because of something lost, but I would take our voice related medical interventions any day of the week.

[@rentourapartment]

Hi Stone, you're arguing against things that I haven't brought up, and suggesting that my view is something it is not, so I'll leave you with good wishes for your singing future. :-)