The Story

Why this exists — and what it cost to get here.

I am going to tell you something that most people in the vocal world would consider either impossible or irrelevant. I am a basso profondo. My functional range runs from G1 — approximately 48 Hz, below the lowest string of a double bass — to Ges5, at 755 Hz, well into soprano territory. I do not switch registers to get there. There are no registers. There is only a pressure system — and when it is balanced, it moves through nearly four octaves the way a continuously variable transmission moves through its gears. Without interruption. Without effort. Without the voice fighting itself.

I know this because I have spent two years measuring it. Spectrograms, waterfall analyses, Praat acoustic measurements across the full range — every tone documented with date, time, and complete voice report. The data is unambiguous. And I know it because I spent twenty years before that building it — alone, without a teacher, without institutional validation, without anyone to tell me I was going in the right direction.

This is that story.

The stage

I spent more than twenty years as a professional bass singer. Fifty leading operatic roles — Gremin, Mephistopheles, Vodnik, Dulcamara, among others. I worked under conditions where the voice does not have the luxury to fail. Three-hour performances in large halls, over full orchestras, without amplification. Night after night, season after season.

During those years I developed a growing certainty that almost everything I had been taught about the voice was wrong. Not wrong in its intentions — wrong in its direction. The pedagogical tradition I had inherited was built on the idea that a voice is constructed from the bottom up. Low notes first. Then middle register. Then the passage — the passaggio — navigated carefully with covering and mechanical adjustment. Then the upper register, developed separately, joined to the rest through technique.

I tried this. It produced the results it always produces — a voice divided against itself, compensating at every transition, gradually accumulating the tension that eventually limits and ends careers. I watched it happen to singers I respected. I felt it beginning to happen to me.

So I stopped. Not performing — singing the way I had been taught. And I started again, from a different direction.

The laboratory

The principle I arrived at is simple, though its implications are not. The voice does not build upward from its fundamental. It resonates outward from a harmonic centre that already contains everything.

For a basso profondo, that centre sits around Fis5 — 755 Hz, nearly four octaves above the lowest functional pitch. Fis5 is the sixteenth harmonic of G1. It is physically present — acoustically, measurably present — every time the lowest note sounds. The Singer’s Formant, the resonant energy between 2,500 and 3,500 Hz that allows a trained voice to carry over a full orchestra without amplification, is not something you build from below. It is what the vocal tract amplifies when it is oriented around the harmonic space that Fis5 inhabits.

When I oriented my practice around this principle — when I approached the full range from above rather than building toward it from below — the compensations began to disappear. Not overnight. Over years. Layer by layer. Each one revealing more of what had always been there.

The measurements confirmed what the practice had found. On H4 pianissimo, Praat autocorrelation reached 0.9996. HNR — harmonics-to-noise ratio, the acoustic measure of how cleanly a voice vibrates — reached 34.886 dB on the same measurement. On D5 pianissimo, HNR reached 36.881 dB. And on C5, measured twice across two days, HNR reached 41.080 dB — a value I have not found documented in published literature for a live singing voice.

These are not performance numbers. They were measured at home, before breakfast, on a Neumann U87 into an RME Fireface UFX. They are the numbers of a system that has nothing to hide because there is nothing hidden.

In May 2026, something occurred that I had not anticipated. Praat’s formant analysis — a mathematical algorithm that identifies resonant peaks in the vocal spectrum — identified the Singer’s Formant as a discrete formant at 3,159 Hz on Cis5, and again at 3,197 Hz on Ges5 ten days later. For a bass voice operating at these frequencies, this is acoustically anomalous. The algorithm does not interpret. It measures. It found a stable, distinct resonant peak at exactly the frequency where the Singer’s Formant should be — because it was there.

I came to singing through acting. For me, the voice was never an end in itself — it was the instrument through which dramatic truth became audible. A voice that is fighting its own mechanical compensations cannot respond freely to dramatic impulse. It is occupied with its own survival. The microexpression, the whispered threat, the pianissimo that carries grief across a two-thousand-seat hall — these require a system that is not spending any of its resources on holding itself together.

This is why pianissimo matters more than forte for me. Forte forgives. Pianissimo reveals. The measurements at piano and pianissimo are consistently higher than at forte — not because forte is strained, but because the absence of pressure exposes the architecture underneath. The system either stands on its own or it does not. At 36.881 dB HNR on D5 pianissimo, it stands.

The single-variable experiment

In the course of these measurements, one experiment produced results I consider directly relevant to the wider practice of vocal pedagogy.

On the same evening, using the same microphone and interface, I sang the same pitch twice — once with my natural tongue position, once imitating the retracted tongue position characteristic of a particular style of Germanic operatic singing. Everything else remained constant: the vowel, the dynamic, the pitch, the room.

Natural tongue position: HNR 22.207 dB. Retracted tongue position: HNR 14.130 dB.

Eight decibels. One variable. Every three decibels of HNR represents approximately a doubling of the ratio of harmonic energy to noise. Eight decibels is not a marginal difference. It is a structural one.

The retracted tongue narrows the pharyngeal space. The narrowed pharynx disrupts the epilaryngeal configuration that produces the Singer’s Formant. The result sounds darker. It sounds operatic. But the acoustic architecture is compromised — measurably, reproducibly, in a single session with a single variable changed.

This is not a criticism of any particular tradition. It is a description of what happens physically when the pharyngeal space is reduced. The physics does not negotiate with aesthetics.

The decision

I stepped away from the stage not because the voice failed. It did not fail. I stepped away because I wanted to be present for my children during years that do not come back. It was not a difficult decision. It was the right one.

What I did not expect was what stepping away would clarify. Without the pressure of performance, without the compromises that professional singing demands, the work became purely itself. The laboratory became the stage. The spectrogram became the audience — more honest than any of them, because it does not applaud and it does not lie.

Over twenty years of working alone, I arrived at a methodology that I have never seen described anywhere in pedagogical literature. Not because it is secret — because it follows the physics rather than the tradition, and the tradition has been moving in the wrong direction for long enough that the physics has been largely forgotten.

Why now

I am fifty-one years old. I have behind me everything I described above, and ahead of me whatever time remains to pass it on. I am not interested in building an institution or a brand. I am interested in finding the singers for whom this work will matter — the ones who feel the limitation that technique cannot solve, who know that something is wrong but cannot locate it, who have been told to cover and push and manage and navigate when what they needed was to stop doing all of that and listen to what the physics was already telling them.

The diagnostic session is not a lesson. It is a map — a real-time spectral analysis of where your pressure system loses its balance, and a direct experience of what balance feels like when it is restored. Most singers feel the shift within the same forty-five minutes. Not because they learned something new. Because they stopped doing something that was in the way.

The cohort is for those who want to stabilise what the diagnostic reveals. Five singers at a time. Eight weeks. Specific work on specific imbalances, visible in data at every session. Not generic exercises. Not one-size-fits-all technique. The precise removal of what is blocking what was always there.

I spent twenty years working in isolation on a path that the conventional vocal world would have called unnecessary at best and dangerous at worst. The data says otherwise. The spectrograms say otherwise. The Praat analysis says otherwise. And fifty roles over twenty years of professional performance say otherwise.

I am not asking anyone to take my word for it. I am asking them to look at the data — their own data, measured in real time — and decide for themselves.

That is what the diagnostic session is for.

Ondrej Mraz

Bass singer · Vocal coach · singinglightly.com