I always start this lesson with a glass of water. Not literally, most days, but the image does the real teaching. Pour water into a tall thin glass and it becomes a tall thin column. Pour the same water into a wide shallow bowl and it spreads out flat and wide. The water doesn’t decide its own shape. The container does.
Your breath works exactly the same way. The air coming up from your lungs is just air, formless, until it hits the shape you’ve built for it on the way out. That shape is your mouth, and it’s not one fixed thing. It’s a container you’re actively building and rebuilding with every phrase. The moving parts are your articulators: your jaw, your tongue, your lips, your cheeks, your soft palate. Every one of them is a wall of that container, and every one of them has to be willing to move. And the container doesn’t stop at your lips either. It runs all the way down through your throat, your pharynx, the space above your vocal folds. All of it shapes the sound before it ever leaves your body.
In everyday talking, you never really test this. Speech moves fast, and phoneticians who study this for a living have a name for what happens because of that speed: undershoot. Your tongue and jaw are still traveling toward the full shape of a vowel when the next consonant already needs them somewhere else, so you only ever arrive partway there before you’re pulled off toward the next sound. Nobody notices, because the ear is used to filling in the rest and speech was never asking for a finished shape in the first place. It’s asking for something recognizable enough to get out of the way fast.
Singing takes that excuse away completely. When you hold a note across a sustained pitch, sometimes for two beats, sometimes for eight, sometimes for a whole phrase, there’s no next syllable yanking your jaw off course before the shape has arrived. The vowel gets all the time it needs to fully form. Whatever container you actually built when you opened your mouth on that note is the container the sound has to live in for as long as the note lasts, no partial credit, no ear filling in gaps for you. If it’s cramped, the sound stays cramped for the whole ride.
This is why vowel expansion matters so much more in singing than it ever does in speech. Vocal scientists have actually measured it. Two of the resonant frequencies your vocal tract produces do most of the work of telling your ear which vowel it’s hearing, and researchers just call them the first and second formant, F1 and F2. Drop your jaw and lower your tongue, and F1 climbs, because you’ve made the main chamber of your mouth bigger, and a bigger chamber resonates lower which, counterintuitively to a lot of singers, shows up as a higher first formant reading. Push your tongue forward instead, and F2 rises. Put a microphone in front of a trained voice and a spectrogram behind it, and you can watch those frequency bands move a visible amount the instant the jaw drops a centimeter. The shape of your mouth doesn’t just look a certain way, it’s tuning your voice in a way you can literally graph. Open the space and you give those frequencies room to build against each other and reinforce. Close it down and you cut them off before they’ve had the chance to add up to anything.
Trained singers who can be heard over a full orchestra with no microphone aren’t doing that through sheer volume. Researchers studying opera singers found something specific happening acoustically: several of the higher formants, the third, fourth, and fifth, cluster together into one strong spike around three thousand hertz. Voice teachers call the effect ring, or squillo, or ping. It works because an orchestra puts out a lot of energy in the lower frequencies and comparatively little up around three thousand hertz, so a voice that produces a strong peak exactly there cuts straight through the texture instead of getting buried under it. That clustering comes from tuning further down that same container, a narrower point right above the vocal folds paired with a wide open throat behind it, which is really just the same principle as the mouth shape we started with, applied one level deeper. The whole tube matters, not only the part you can see in a mirror.
I like to show the mouth-shape piece of this with something as ordinary as the phrase “I’m with you.” Said the way you’d say it to a friend on the street, it barely takes any shape at all. The “you” collapses into something closer to a clipped little grunt, and nobody notices because speech never asks it to hold still. Sing it that way, tight jaw, barely parted lips, and you will feel your own throat working against you. But open it. Let the “you” stretch into something closer to “ya,” let your jaw drop and your lips round around it. The same three words suddenly carry. It’s still recognizably “you,” the vowel hasn’t turned into something else, it’s just been given the room it needed to actually ring.
When you clamp down instead, a few things happen at once, and none of them help you. Your articulators stop moving, so the container stops adjusting to what the sound actually needs from one note to the next. Your breath backs up against that narrow opening, and the muscles around your jaw and throat tense up trying to push past it, which is its own kind of choke hold on the airflow. The formants that were supposed to spread out and reinforce each other get squeezed together instead, muddying the vowel and flattening exactly the frequency range that would have let the sound carry to the back of a room. The resonance, the very thing that was supposed to fill the room, gets strangled right at the source. You end up working harder for less sound, and less of the sound that would have actually traveled. That’s the real problem with a closed mouth. It’s not wrong exactly. It’s just expensive.
When I’m working with a student on this, I usually don’t start by talking about formants or airflow at all. I just get them to exaggerate. Ridiculously open jaw, lips doing far more than they’d ever do in conversation, vowels stretched almost embarrassingly wide. It feels like too much at first, like overacting with your own face. But that exaggeration is what teaches the muscles the range they actually have, because most people are working with maybe a third of the space available to them and calling that their voice. Once the body learns what open really feels like, you can dial it back down to something that looks completely natural on stage and still carries to the back row.
So before you worry about pitch, or breath support, or any of the more technical pieces, look at your mouth. Ask whether it’s actually shaping the sound or just letting it leak out around the edges. The container comes first. Everything you pour into it takes that shape.
Further reading: The Singer’s Formant, National Center for Voice and Speech. Formant, Voice Science.