Baby Talk and Language Development

Baby talk can mean two things. One is parentese, real words spoken in a higher, slower, sing-song voice, and group studies link it with later language. The other is made-up words such as “din-din,” which the American Speech-Language-Hearing Association says are okay to use sometimes, as long as the adult word comes too.

What you’ll learn

• The difference between parentese and made-up baby words
• What group studies and coaching trials have found about parentese
• What health organizations advise about family words like “din-din”
• How talking with a child shifts as the child grows

Is Baby Talk Good or Bad for Language Development?

Whether baby talk helps or hinders depends on which kind is meant, because the phrase covers two separate things. One is a way of speaking: a higher pitch, a slower pace and a sing-song rise and fall, used with ordinary words and full sentences. Researchers call this parentese, or infant-directed speech. The other is a set of words: made-up or family words, such as “din-din” for dinner, used in place of the adult word.

The worry itself is old. A 2020 paper in the Proceedings of the National Academy of Sciences, by Ferjan Ramírez, Lytle and Kuhl, notes that early writers on the subject warned this way of talking might damage children’s language development, and that later work found parentese to be fully grammatical, with words for objects and actions a baby can see.

Part of the confusion is the name. When anthropologists and linguists first described this way of speaking in the 1960s, they called it “baby talk,” and a 2017 paper in Frontiers in Psychology notes that it is still commonly called either baby talk or parentese. So a warning about baby talk and a study in its favor may not be about the same thing at all.

From clinical practice

The high, slow voice that comes so easily with a baby is real words said warmly and stretched out. That voice is what the parentese studies measure. Made-up words such as “din-din” are a separate question, with advice of their own.

Parents tend to worry about the voice as baby talk. Of the two, it is the part the research speaks to.

What Parentese Sounds Like, and Why Babies Listen

The 2020 paper describes parentese as a near-universal speaking style marked by higher pitch, slower tempo and exaggerated intonation, documented in speech to young children in many countries. According to the 2017 paper, its sounds are clearer, longer and more distinct from one another than in ordinary adult speech. The 2020 authors also summarize earlier acoustic studies showing that its vowels are stretched out and pulled further apart across languages, which makes it easier for babies to hear the differences between speech sounds.

The American Academy of Pediatrics (AAP) sets out the same shift from the parent’s side: raising the pitch, slowing down, exaggerating syllables and widening the eyes and mouth, which it says captures a baby’s attention. The AAP also notes that babies favor the human voice over other sounds and like high-pitched voices in general.

What Studies of Parentese Have Found

A study run across 67 laboratories on four continents, published by the ManyBabies Consortium in 2020, confirmed that babies as a group prefer listening to infant-directed speech over speech aimed at adults, though the preference was smaller than earlier studies had reported. It was stronger in older babies and when the speech matched a baby’s own language and dialect.

A 2015 review in Current Directions in Psychological Science concludes that infant-directed speech draws babies’ attention to language, encourages social interaction between babies and caregivers, and helps babies pick up features of their native language. The same review points to newer research on the importance of the caregiver responding to the baby, not only to the sound of the voice.

Studies That Followed Families Over Time

In a 2014 study in Developmental Science, researchers recorded the home language of 26 babies at 11 and 14 months. Parentese heard one-on-one was linked with more speech from the babies at the time and more words at 24 months, and both the one-on-one setting and the style of speech were linked with the children’s later language. A 2017 follow-up in Frontiers in Psychology, with 30 children at 33 months, found that the link with one-on-one parentese in infancy still held.

A 2024 study in the Journal of Child Language followed 44 English-speaking children and found that parents’ steady, frequent use of parentese in infancy was linked with the variety of words the children used, the length of their sentences and how often they took turns in conversation when they started kindergarten.

All three describe small groups and show a link, not a cause. The 2017 authors caution that the link may run both ways, since caregivers may use more parentese with children who are already more talkative or social.

Trials That Coached Parents

Two randomized controlled trials, in which families were assigned by chance to coaching or no coaching, tested coaching directly. In the first, published in Developmental Science in 2019, parents coached when their babies were 6 and 10 months old used more child-directed speech and more parentese. As a group, their babies’ babbling grew more, and at 14 months those babies produced more words than babies in the comparison group.

The second was the 2020 trial by Ferjan Ramírez, Lytle and Kuhl, with 71 families, in which parents were coached when their babies were 6, 10 and 14 months old. Coaching increased parents’ use of parentese and of back-and-forth turn-taking through 18 months, and those increases went along with the children’s language growth over the same period and their language at 18 months. Both trials coached a package, the amount of child-directed talk, turn-taking and parentese together, so neither tests the voice on its own.

The 2020 authors add that it is not yet known whether their findings apply to bilingual or multilingual families or to children with clinical conditions, and the 2024 study included only English-speaking children, so these results describe the families studied and no more.

Made-Up Words and Family Words for Things

Made-up words are a separate question from the voice, and the guidance on them comes from health organizations, not from the parentese studies. The American Speech-Language-Hearing Association (ASHA) puts it this way: it’s okay to use baby talk sometimes, as long as you also use the adult word, giving the example of “din-din” alongside “dinner.” That tip sits in ASHA’s list of activities for 2 to 4 years, not in its advice for babies, and the same list advises speaking clearly and modeling good speech.

The AAP’s advice on a child’s own invented words points the same way. When a child makes up a word, such as “mog” for milk, the AAP advises treating it with all the respect of a real word while answering with the adult one, “milk.” It also recommends keeping language simple and specific, with consistent names for things, so that a pet called a cat today is not a kitty tomorrow. Taken together, the advice from both is to keep the adult word in the conversation alongside any family word.

Talking With Your Baby Day to Day

For the first year, ASHA’s advice is to talk to your baby, imitate the sounds your baby makes and use the language or languages your family is most at ease with. The 2015 review stresses responsiveness, which in practice means answering whatever your baby offers, a sound, a look or a gesture, and then give your baby room to answer back.

Copying a coo and adding a simple word or phrase does both at once, and there is more on reading those early signals in how babies communicate before words. How babies come to say mama follows one first word from babbled sound to a name for a parent.

How Talk Changes as a Child Grows Older

In the 2017 study of 30 children, caregivers used less parentese and more ordinary speech one-on-one as their children got older. By 33 months, the kind of talk linked with the children’s word use at the time was ordinary one-on-one speech, not parentese.

The AAP’s advice points the same way, suggesting that parents mix adult language into their conversations with a baby. No age for moving on comes from ASHA or the AAP from the sing-song voice, and in the 2017 families the change came as the children grew.

These studies describe groups of families, so they do not explain one child’s language or the cause of a delay in a particular child. The amount of talk around a child is its own question, taken up in language exposure and speech delay, and the speech and language development pages cover how talking grows more broadly.

Common Questions From Parents

Is Baby Talk Bad for My Baby’s Language?

Not if baby talk means the high, slow, sing-song voice used with real words, which group studies link with more words later on. Made-up words are a separate question, and ASHA’s view is that they’re okay now and then if the adult word is there as well.

What Is Parentese?

Parentese is the name researchers use for a speaking style found in nearly every culture: higher pitch, slower pace and an exaggerated rise and fall, with vowels stretched out. It uses real words and full grammar, which sets it apart from made-up baby words.

Are Made-Up Baby Words Harmful?

ASHA’s guidance, in its list for 2- to 4-year-olds, is that baby talk is okay on occasion as long as the adult word comes alongside it, as in “din-din” and “dinner.” The same list asks parents to speak clearly and model good speech.

When Should I Stop Talking to My Child in a Baby Voice?

Neither ASHA nor the AAP gives an age. In a 2017 study, caregivers shifted from parentese toward everyday speech as their children grew, and by 33 months everyday one-on-one talk was what went along with word use. The AAP suggests mixing adult language into conversations.

Is It Okay to Use Our Family’s Own Words for Things?

Family words can sit alongside the adult ones. If your child coins a word, like “mog” for milk, the AAP’s advice is to treat it as a real word and reply with “milk,” and to keep each thing’s name the same.

Reviewed by Monica d’Errico-Franklin, M.S., CCC-SLP

More about Monica