There is a quiet kind of guilt that comes with a toddler who isn't talking yet. It sends you to the internet at eleven at night, where you find a confident list: blow bubbles, drink thick smoothies through a straw, offer chewy jerky, make silly faces in the mirror. The logic feels airtight. Speech comes from the mouth. The mouth is made of muscles. Train the muscles, get the words.
It is one of the most intuitive ideas in all of child development. It is also one of the least supported. And the gap between what feels obvious and what the evidence shows is worth understanding, because it can save you months of effort pointed in the wrong direction.
Why the muscle theory feels so right
We talk about speech as if it were physical fitness. People say a child has a "weak mouth" or a "lazy tongue." Therapists in the past reached for the same picture: if a child's speech is unclear, strengthen the lips, tongue, and jaw with drills that look a lot like a gym routine.
These drills are called non-speech oral motor exercises, often shortened to NSOMEs. They include blowing through straws or whistles, pushing the tongue against a spoon, puffing out the cheeks, and chewing resistant foods. Notice what they share: none of them involves making a speech sound.
That last fact is the whole problem.
Speech is a skill, not a strength
Saying the word "ball" does not require much force. The muscles of the lips and tongue are used at a tiny fraction of their capacity. What the word requires is timing and coordination: lips closing at exactly the right moment, the jaw dropping, the voice starting a beat later, all in a sequence your brain has learned.
That makes speech much more like playing a piece on the piano than lifting a weight. Nobody improves their piano playing by squeezing a stress ball, even though both use the hands. Motor learning researchers describe this as task specificity: the brain gets better at the movement you actually practice, and the benefit transfers poorly to movements that merely look related.
The research on oral motor exercises reflects this. Reviews of the evidence, including a Cochrane review of non-speech oral motor treatment for children with speech sound disorders, have found insufficient evidence that these exercises improve speech. The American Speech-Language-Hearing Association has likewise noted that the evidence for using non-speech oral motor exercises to change speech is lacking. That does not mean the drills are harmful. It means that, as a way of producing words, they have not earned their reputation.
What the mouth is actually doing when a baby babbles
Here is where the real science gets interesting. Researchers Peter MacNeilage and Barbara Davis proposed what is known as the frame/content theory. In early babbling, the jaw does most of the work. It opens and closes rhythmically, and that simple up-and-down motion becomes the "frame" of a syllable. The tongue and lips mostly ride along, and only later learn to move independently inside that frame, the "content."
This is why a baby's first syllables are things like "ba," "da," and "ma." The jaw swings; the sound falls out. Over the second year, a child gradually separates tongue and lip movements from the jaw so they can produce more varied sounds.
Notice what drives that progress. It isn't a muscle getting stronger from chewing. It is a brain getting better at controlling movement, and that happens through hearing speech, producing sounds, and hearing the results come back. Babbling is practice. Early words are practice. The workout is the talking itself.
So are chewy foods and straws useless?
Not at all. They just aren't speech therapy, and they don't need to carry that weight.
Chewing a variety of textures is good for feeding development, and feeding is a legitimate skill in its own right. Drinking from an open cup or a straw is a lovely milestone. Blowing bubbles is a joyful game. If your toddler has genuine feeding difficulties, such as gagging on lumpy food, pocketing food, or refusing whole categories of texture, a feeding specialist can help, and that is a separate question from speech.
The mistake is the transfer: assuming that because the mouth is busy, words are being built. A child who chews beautifully may still be a late talker, and a child who lived on purees may talk fluently. Feeding skills and speech skills share anatomy, but they are learned largely separately.
There is also a real exception worth knowing. Some children have a motor speech disorder such as childhood apraxia of speech, where the difficulty is in planning and sequencing speech movements. Red flags can include very little babbling as a baby, few consonant sounds, and inconsistent errors on the same word. Treatment for that is still not a set of chewing exercises. Evidence-based approaches for apraxia use speech practice itself, with many repetitions of real sounds and words, guided by a speech-language pathologist.
What does move the needle
If the mouth is a skill instrument, then the best training is playing it, in the context of real communication. A few things the research consistently supports:
- Hearing lots of responsive, back-and-forth speech. Children learn sounds from sounds, not from drills.
- Producing sounds in play. Games with vehicle noises, animal sounds, and silly syllables give the jaw and lips low-pressure practice on real speech movements.
- Seeing mouths. Face-to-face time lets a child watch how sounds are shaped.
- Hearing words in short, clear, repeated form, so the movements have a stable target to aim at.
Notice that none of this requires equipment, and none of it feels like a chore.
Your next moves
- Swap the drill for a sound game today. Pick two easy sounds your toddler already makes, like "ba" or "mama," and pass them back and forth for two minutes during play. You are practicing real speech movements.
- Get face to face. At one meal or diaper change, get down at eye level so your toddler can see your lips shape a word like "pop," "up," or "more."
- Keep feeding and speech in separate boxes. Offer varied textures at meals because it helps eating. Skip the guilt if it doesn't change talking.
- Write down what you hear. For three days, jot down the consonant sounds your child uses (b, m, d, n, p, t, h are common early ones) and any words, even approximations. That list is exactly what a speech-language pathologist will want.
- Book a screening if you see red flags. If your child has very little babbling, few consonants, trouble with feeding, or is not using words by around 18 to 24 months, ask your pediatrician for a speech-language evaluation. You don't need to wait for permission.
The relief inside the research
There is something freeing in knowing the mouth is not a muscle you have to build. You are not behind because you skipped the straw smoothies. Your toddler's words are growing out of something far more ordinary and far more available: being talked to, being answered, and having a few unhurried minutes to try a sound and hear it land.
That is the idea behind Acorn. It is a daily three-minute first-words session for one- to three-year-olds, built around short, clear, repeatable words and real back-and-forth practice rather than gadgets or drills, with no ads and no upsells. If you'd like a small, steady rhythm to hold onto, you can look around at acorn.lumenlabs.works. And if you never do, the sound game you play at dinner tonight is already the right idea.