The science of overtone singing

What happens inside your voice when a harmonic appears

Overtone singing sounds like a mystery: two pitches from one voice. But the mechanics behind it are knowable, measurable, and once you understand them, profoundly useful for anyone who works with the voice.

This page is for singers, voice teachers, researchers, and anyone curious about the acoustics of the human voice. I’ll walk you through how overtone singing works, what we know from current science, and where the open questions are, including two hypotheses I’ve published to address them.

Every voice is a "chord"

When you sing a single note, you’re actually producing dozens of frequencies simultaneously. Your vocal folds vibrate at a fundamental frequency (the pitch you hear), but they also generate a series of integer multiples above it, the harmonic series: 2x, 3x, 4x, 5x, and so on.

This is not mystical. It’s physics. Every vibrating body produces harmonics. A guitar string does it. A bell does it. Your voice does it.

In ordinary singing and speaking, these harmonics blend together into what you perceive as “timbre”, the quality that makes your voice sound like yours. You don’t hear individual harmonics because your vocal tract shapes them into vowels and consonants, emphasizing some and dampening others.

Overtone singing is the art of making individual harmonics audible.

The vocal tract as a filter

The key to overtone singing lies in the source-filter model of voice production. Your vocal folds are the source, they produce a raw, buzzy signal rich in harmonics. Your vocal tract (throat, mouth, nasal passages) is the filter, it shapes that signal by amplifying certain frequency bands and attenuating others.

These bands of amplification are called formants. When you change the shape of your mouth and tongue to say different vowels, you’re moving formants around. The vowel “ee” has different formant positions than “oo”, that’s why they sound different, even at the same pitch.

In overtone singing, you learn to position your formants with extreme precision, and more precisely you merge two different resonant frequencies, concentrating their coupled energy on a single harmonic. The result: that harmonic rises above the rest and becomes audible as a distinct, whistling tone above your drone.

The spectrogram — making the invisible visible

A spectrogram is a visual representation of sound that shows frequency (vertical axis), time (horizontal axis), and intensity (brightness). For overtone singing, it’s an indispensable tool.

On a spectrogram, a normal sung note appears as a stack of horizontal lines, the fundamental at the bottom, harmonics above it, gradually fading. When an overtone singer isolates a harmonic, one of those lines suddenly blazes bright while the others dim. You can literally see the harmonic melody being traced in real time.

I use spectrograms (via VoceVista and Praat) in my teaching and research. They allow you to see what your ear is learning to hear, a feedback loop that accelerates learning enormously.

The "Flute in the Throat" hypothesis

Here’s where current science meets an open question.

The standard acoustic model explains overtone singing through formant tuning: you narrow a formant bandwidth to isolate a harmonic. This works well as a general explanation. But it doesn’t fully account for what researchers have measured in skilled overtone singers, particularly in the sygyt style, where the isolated harmonic is narrower, more intense, and more stable than soft-tissue resonance should allow.

In 2025, I published a hypothesis on Zenodo proposing an additional mechanism: vortex-acoustic lock-in. The idea, drawn from fluid dynamics and aeroelastic research, is that under certain airflow conditions, small vortices formed at constrictions in the vocal tract can synchronize with a resonant frequency, creating a self-sustaining oscillation that acts like a tiny flute embedded within the vocal tract.

Not a flute that replaces the voice. A flute that tunes itself to the voice.

This would explain the extraordinary precision of the overtone in sygyt: it’s not just passive filtering, but an active aeroacoustic mechanism that locks onto and reinforces a specific harmonic.

The hypothesis is currently being discussed with researchers in acoustics and medical imaging, and MRI-based protocols are in preparation to test its predictions.

This is not settled science, it’s a working hypothesis. I share it here because I believe in transparency: showing not just what we know, but where the edges of knowledge are. The most honest thing a teacher can say is “here’s what I think is happening, here’s the evidence so far, and here’s what we still need to test.”

Read the full paper: “Flute in the Throat” on Zenodo →

Overtone singing and the brain.
The theta-gamma hypothesis

The “Flute in the Throat” hypothesis addresses what happens inside the vocal tract. But what happens inside the brain?

In 2025, Wolfgang Saus, together with Seither-Preisler and Schneider, published MEG (magnetoencephalographic) data showing that listening to overtone-rich sound drives a pronounced increase in theta-band cortical activity (4–8 Hz), with a strong right-hemisphere lateralization that accounts for over 80% of the variance in brain activation patterns. In other words: overtone singing doesn’t just sound unusual, it activates the brain in a measurably distinct way.

Building on this finding, I published a second hypothesis on Zenodo in 2026: “Overtone Singing as Natural Theta-Gamma Cross-Frequency Neuromodulation.” The core idea is that active overtone singing (not just listening, but performing) may create a unique neurophysiological state where theta-dominant auditory self-stimulation co-occurs with gamma-band activation from the fine motor control, focused attention, and auditory-motor feedback that the technique demands. This simultaneous dual-band engagement could enhance theta-gamma coupling endogenously, a neural mechanism that supports working memory and whose degradation is among the earliest markers of Alzheimer’s disease and cognitive decline.

If confirmed, overtone singing practice could represent an accessible, non-pharmacological intervention with potential neuroprotective benefits for cognitive aging.

This is a hypothesis, not a clinical claim. I share it here for the same reason I share the “Flute in the Throat” work: because I believe transparency about what we know, what we hypothesize, and what remains to be tested is more honest, and more useful, than either silence or certainty.

Read “Overtone Singing as Natural Theta–Gamma Cross-Frequency Neuromodulation” on Zenodo →

Three Ways of Inquiry

Through my work, I apply three scientific lenses.

The first is evidence-based science: acoustic measurements, spectrographic analysis, the source-filter model, published research. This is the science of “what can be measured and replicated.” It gives us formant frequencies, harmonic ratios, airflow dynamics. It’s rigorous, testable, and essential. Wolfgang Saus’s MEG research on the listener’s brain response to overtone singing is a beautiful example measuring something that practitioners have intuited for centuries, and giving it a neurophysiological foundation.

The second is experience-based science: the knowledge that comes from sustained, attentive practice. This is the science of “what a trained ear and body can perceive that instruments haven’t yet captured.” Experienced overtone singers (across all traditions) describe sensations, internal feedback loops, and perceptual states that acoustic models don’t yet fully explain. The “Flute in the Throat” hypothesis itself was born from this gap: something I consistently felt and heard in my own practice that the standard models didn’t account for.

The third is integrative science: the systematic bridging of measurement and embodied experience through structured phenomenological inquiry. This is the science of “what emerges when we rigorously describe and validate lived perceptions against data.” It uses protocols like empirical phenomenology to map internal sensation, such as laryngeal resonances or perceptual feedback in diphonic singing, making them shareable, testable, and integral to research, as seen in studies of vocal embodiment and expertise.

No single kind replaces the others. Evidence-based science without embodied experience produces descriptions that don’t help you sing. Experience-based science without measurement produces claims that can’t be verified. Integrative science without both remains abstract.

The approach I teach, and the research I pursue, lives at their intersection. I measure what I can, practice what I feel, integrate what validates both, and remain honest about which is which.

Current research in overtone singing

The scientific study of overtone singing has grown significantly since Levin and Edgerton’s landmark 1999 analysis of Tuvan styles. Key contributions include:

This is an active field. The research is moving along two parallel tracks: understanding how overtone singing works acoustically, and understanding what it does neurologically. New imaging technologies and computational models are opening questions that weren’t possible to ask a decade ago. I share updates on this page as the research evolves.

Why the science matters for singers

You don’t need to understand acoustics to sing overtones. But understanding it changes how you learn.

When you know that the overtone emerges because of formant positioning, not because of some ineffable gift, you stop guessing and start adjusting. When you can see on a spectrogram exactly which harmonic you’re amplifying, you get immediate, objective feedback. When you understand the source-filter model, you realize that your entire voice (not just overtone singing) is built on the same principles.

The science doesn’t replace the experience. It illuminates it. And in my teaching, it’s the bridge between “I heard it once but can’t find it again” and “I understand what I did, and I can do it every time.”

Scroll to Top