For littleWords.ai, the goal is not to turn parents into therapists. The goal is to make everyday moments easier to join, easier to repeat, and easier for a child to use in their own way.
I am not a perfect parent. I do not run a Montessori-inspired calm-down corner. My kid eats chicken nuggets four nights a week. When the SLP told me I needed to do daily speech practice with my late-talking three-year-old, I almost laughed in her face.
Daily. With a kid who runs out of the room if I say the word “practice.” Sure.
But she was right about one thing: 30 minutes of therapy a week is not enough. The other 167 hours matter more. So I figured out how to do it in 10-minute chunks, woven into things we were already doing. Here’s what an actual week looked like. Maybe it’ll help you.
The getting-dressed trick that changed mornings
My friend Danielle in Rochester has a four-year-old, Leo, who had about 15 words at his third birthday. She told me once over coffee, “Our SLP said choices are free therapy, and I thought that was the dumbest advice I’d ever gotten. Then it worked.” Her version: she’d hold up two pairs of shoes every morning and just wait. Leo went from pointing silently to saying “boots” in about three weeks. “He said it so clearly one Tuesday that I texted my husband a voice memo from the hallway,” she told me. “I was literally shaking.”
That’s almost exactly what happened at our house, just with shirts instead of shoes.
Monday mornings used to be the worst part of our day. He hated transitions, hated the cold of the room, hated socks. I’d bark instructions and he’d melt down.
Our SLP suggested I turn the routine into a predictable script with built-in choices. So now I sit on the floor with two shirts. Hold them up. “Blue shirt or green shirt?” Wait. Count to five in my head.
The first week he didn’t say anything. He’d just point. I’d model the word back. “Green. You want green.” Then I’d put it on him.
By week three, he was saying “geen” before I finished the sentence. By month two, he was the one asking, “Blue or green?” and then choosing.
The trick: I’m not asking a quiz question. I’m offering a structured invitation, and I’m okay with no answer at all. There’s research backing this up. A 2006 study published in the Journal of Speech, Language, and Hearing Research found that “expectant pausing,” where the adult creates a clear communicative opening and then waits, significantly increased spontaneous verbal attempts in toddlers with language delays compared to direct prompting alone (Yoder & Warren, 2006). The wait time is the magic. Five seconds feels like an eternity when you’re standing in a cold bedroom holding two shirts, but it gives the child’s brain room to formulate and attempt. Our SLP told me most parents wait about 1.5 seconds before jumping in. The kids who need the most time get the least of it.
I started timing myself with a mental count. One-Mississippi, two-Mississippi, all the way to five. Some mornings he’d answer at three. Some mornings he’d answer at six. Some mornings he wouldn’t answer at all, and I’d model and move on. All three of those are fine.
Breakfast, and why the cereal lives on the top shelf
He loves a specific cereal. I keep it on the top shelf, out of reach. He has to ask. Not because I’m withholding to torture him, but because the cereal is a built-in reason to communicate.
This strategy is called “environmental arrangement” in the speech therapy world. You set up the physical space so that natural communication becomes necessary. The American Speech-Language-Hearing Association (ASHA) lists it as a core parent coaching technique for early intervention: placing desired items in view but out of reach, giving small portions so the child has repeated opportunities to request, and pausing before fulfilling a need the child clearly has (ASHA, 2023). The point is not to frustrate the child. The point is to create gentle, repeated chances to use language for its actual purpose, which is getting something you want.
When he was new to words, he’d point and grunt. I’d say, “Cereal? You want cereal?” and pour it. I gave him the word. I didn’t make him perform.
When he had a few words, I’d say, “Cereal. You want some?” and wait. If he said “ceeo,” I’d celebrate and pour.
When he had more words, I’d say, “Tell me what you want.” And he’d say, “Cereal please mama.” And I’d cry a little bit, internally.
Same routine. Different rung on the ladder. That’s the whole game.
A friend of mine tried this with bananas and almost gave up in the first week because her son just screamed and pointed. She felt cruel. But her SLP reminded her that the goal at the beginning is not the word. The goal is shared attention, the moment where the child looks at you, acknowledges that you are the path to the banana, and engages. The word comes later. The looking-at-you part is the foundation. Once she reframed success as “he looked at me and pointed” instead of “he said banana,” the whole thing got easier for both of them.
The car is a speech clinic (and it’s free)
I used to spend the car ride scrolling at red lights. Now I sing.
Songs are the secret weapon of speech practice. They’re predictable. They have a melodic structure that helps language stick. They tolerate weird pronunciation. And they don’t feel like work. Think of it like muscle memory for the mouth: the melody gives the brain a scaffold to hang the words on, the way learning lyrics is easier than learning a monologue.
There’s a neurological reason for this. Research from the Auditory Neuroscience Laboratory at Northwestern University has shown that musical training and melodic exposure strengthen the same neural pathways used for speech processing and phonological awareness (Kraus & Chandrasekaran, 2010). Singing slows down the rate of speech, exaggerates vowel sounds, and creates a rhythmic framework that makes individual words more salient and easier to imitate. For a child who is struggling to parse the rapid stream of adult conversation, a song is like subtitles for the ears.
We have a rotation of about six songs. “Wheels on the Bus.” “Old MacDonald.” “Twinkle Twinkle.” Whatever your kid likes. I sing the line, then leave the last word hanging.
“The wheels on the bus go round and…”
Pause. Wait.
The first month, silence. Then one day, from the back seat, a tiny voice: “round.” I almost drove off the road.
Now we do “fill in the blank” songs every morning. He’s started doing it to me. I’ll be humming and he’ll yell from the back, “MAMA YOU MISSED A WORD.” Five minutes a day. Free.
One variation that worked well for us: I started making up songs about what we were doing. “We are going to the stoooore, to the stoooore, to the stoooore.” Terrible lyrics. No Grammy potential. But he started filling in the place name before I could. “PARK!” he’d shout, because he knew we were going to the park and he wanted to correct me. Turns out, being wrong on purpose is an incredibly effective speech strategy. Kids who won’t perform on command will absolutely correct you when you say something ridiculous.
Snack time is the most underrated language session in the house
The kid is captive in a high chair. There’s a desirable item right in front of them. They have to communicate to get it. It’s basically a perfectly designed therapy scenario that smells like apple slices.
I cut up an apple. Put one slice on his tray. Wait. He eats it. Another slice. Wait. Eat. Slice. Wait. Eat.
After a few rounds, he started looking up at me between bites. That’s the moment. That’s where you build language. I’d say, “More apple?” and pause. He’d nod, eventually point, eventually say “moh.” Now he says “more apple please” without prompting, and I beam like an idiot every single time.
A 2019 meta-analysis in Language, Speech, and Hearing Services in Schools found that parent-implemented language interventions, especially those embedded in daily routines like mealtimes, produced effect sizes comparable to clinician-delivered therapy for children under four with expressive language delays (Roberts & Kaiser, 2011, updated in Hampton & Kaiser, 2016). The key variable was not the parent’s level of training. It was frequency of practice and responsiveness to the child’s communication attempts. In other words, showing up imperfectly every day at snack time may do as much as a perfect clinic session once a week.
The high chair creates something therapists call a “communication temptation.” The child wants something, the adult has it, and the physical setup naturally requires some form of exchange. You can do the same thing with Play-Doh (give a tiny ball, wait for a request for more), bubbles (blow one round, then hold the wand and wait), or stickers (peel one, pause before handing it over). Any situation where you control the supply and the child controls the demand is a language opportunity.
Bath time, where body parts click fastest
Bath time is a goldmine because the kid is happy, the toys are limited, and the language is naturally repetitive. “Pour. Splash. Wet. Cold. Bubbles.”
We had a bath song. We had a bath script. “Wash the toes. Wash the knees. Wash the belly.” Body part words came in fast because they were tied to a physical sensation and a routine. He learned “elbow” before he learned “spoon.” Which makes sense if you think about it: nobody tickles your spoon.
This connects to something called “embodied cognition,” the idea that words learned through physical experience are encoded more deeply than words learned in the abstract. When your child hears “elbow” while you’re gently scrubbing their elbow, they’re getting the auditory input, the tactile sensation, and the visual cue all at once. That multi-sensory layering gives the word more hooks in memory. It’s the same reason kids learn “hot” fast when they touch a warm mug and you say the word with urgency. Context, sensation, and language fuse together.
Don’t underestimate the toys, either. I had three rubber animals. I’d hold one up. “Who’s this?” Wait. If silence, model. “Duck.” If he tried, expand. “Yellow duck!” If he nailed it, high-five and move on.
One thing I wish I’d known earlier: narrating what you’re doing is different from testing what the child knows. “I’m pouring water on the duck. Splash! The duck is wet now” is narration. “What’s this? What color is this? What sound does this make?” is a quiz. Kids with language delays often shut down during quizzes because they sense the performance pressure. Narration gives them all the same vocabulary input without the demand. Our SLP called it “sportscasting,” and once I shifted to that mode during bath time, he relaxed and, paradoxically, started volunteering more words on his own.
Friday: the tool I use when I’m out of gas
Here’s the thing: by Friday I am tired. I have done a week of “intentional language modeling” and my brain is mush. My kid still needs practice. We need a tool that doesn’t require me to be on.
This is where we use LittleWords.ai. It’s an AI speech companion app built by another parent of an autistic kid, designed for low-pressure expressive language practice. The character is named Buddy. My son lies on the rug with the tablet for about 10 minutes and “talks” to Buddy about whatever he wants (dinosaurs that day, usually). Buddy responds, expands, asks gentle follow-up questions, and waits for his response.
It is not a substitute for me, or for his SLP, or for play with peers. It’s a Friday afternoon pressure valve for both of us. And weirdly, the things he tells Buddy sometimes show up in our conversations the next day, like he’s been rehearsing on his own terms with a character who never gets impatient.
What I appreciate about it is the pacing. Real conversations move fast. Adults talk over kids, interrupt, finish their sentences, redirect. Buddy doesn’t do any of that. It waits the way our SLP told me to wait, except it never forgets, never gets antsy, never glances at a phone. For a kid who needs extra processing time, that patience is a big deal. It’s not magic. It’s just consistent, and on the days when I can’t be, it fills a gap.
What 60 hours of this taught me
Ten minutes a day adds up to about 60 hours a year. That’s roughly twice as much as private SLP sessions deliver. And it happens in moments that already exist in your day, without anyone sitting in a chair staring at flashcards.
You don’t need a curriculum. You don’t need a perfectly decorated playroom. You need:
- A few high-motivation moments built in (snack, bath, getting dressed)
- A willingness to wait five seconds after you say something (this is harder than it sounds)
- The discipline to model the word instead of demanding it
- A song rotation
- A small handful of tools you trust
The biggest mindset shift for me was realizing that “speech practice” wasn’t a calendar block. It was an orientation. Once I stopped trying to find the right hour and just paid attention to the moments already in front of me, my kid started talking more in ways I never could have scripted or planned.
My honest, slightly opinionated take: the parents I know who struggle most with this aren’t lazy or uninvolved. They’re overthinking it. They’re reading about PROMPT therapy and Hanen and the Lidcombe Program and getting paralyzed by the sense that they need to do it “right.” You don’t. You need to do it at all, badly, in your pajamas, with cereal leverage.
You can do this. Ten minutes. Three times today. Start tonight.
Frequently Asked Questions
My child is nonverbal. Does any of this still apply?
Yes. Every strategy here works at the pre-verbal level. The goal at that stage is not words. It’s intentional communication: eye contact, pointing, reaching, gesturing, vocalizing. When I describe waiting for a word at snack time, the earliest version of that is waiting for a look or a reach. You model the word (“apple”) and honor whatever response the child gives. Communication comes before speech, and building that loop of “I express a need, someone responds” is the foundation that words eventually grow from. If your child uses AAC (augmentative and alternative communication) like a picture board or a speech-generating device, these routines work just as well. The cereal still goes on the top shelf. The choice between two shirts still happens. The modality changes, but the structure stays.
How do I know if my child actually needs intervention or is just a “late talker”?
This is the question that kept me up at night. The honest answer: you can’t always tell early on, and neither can professionals with 100% certainty. But there are evidence-based red flags. If your child has fewer than 50 words by 24 months, is not combining two words by age two, doesn’t seem to understand simple instructions (“get your shoes”), or has lost words they previously used, talk to your pediatrician and request a speech-language evaluation. The “wait and see” approach has fallen out of favor among most SLPs because early intervention consistently produces better outcomes than delayed intervention, even if some late talkers do catch up on their own (Hoff, 2013). Getting evaluated does not commit you to anything. It gives you information.
What if my partner or co-parent isn’t on board with these strategies?
This came up in our house. My husband thought I was being weird about the cereal shelf. He’d just hand it over. It created inconsistency, which was frustrating. What helped was having our SLP explain the rationale directly to him during a session. Hearing “environmental arrangement creates natural practice opportunities” from a professional landed differently than hearing “stop giving him the cereal” from me. If your co-parent can’t attend a session, ask your SLP if they can send a short written summary or even a two-minute voice note explaining the “why” behind each strategy. Most people are willing to try something if they understand the reasoning, especially when they see it working.
Is screen time during speech practice a problem?
It depends on the screen time. Passive video watching, where the child stares and the screen talks at them, does not build expressive language. Research from the American Academy of Pediatrics consistently shows that passive screen exposure has a neutral-to-negative effect on language development in children under five (AAP, 2016). But interactive, responsive screen use is a different category. An app that waits for the child’s input, responds to what they say, and adapts its pacing is functioning more like a conversational partner than a television. That’s the distinction. I still limit total screen time. But I don’t count a 10-minute interactive language session the same way I count 10 minutes of YouTube autoplay.
My child only wants to talk about one thing (trains, dinosaurs, etc.). Should I redirect?
No. Ride that obsession for everything it’s worth. A child who is fixated on trains will learn more language through trains than through any “balanced” curriculum you try to impose. Trains give you colors, numbers, spatial words (over, under, through, around), action words (go, stop, fast, slow, crash), and social language (“Thomas is sad because Percy left”). If your child has a topic they’ll talk about willingly, that topic is your entire speech curriculum for now. You can expand from there once the language flows, but trying to force variety before the child has confidence in any topic at all just creates resistance.
How long before I should expect to see progress?
This varies so much that any specific timeline would be misleading. Some kids have a sudden word explosion after weeks of what looks like nothing. Others add words slowly and steadily over months. What I watched for was not word count but communication growth: Was he making more eye contact? Was he pointing more? Was he vocalizing with more variety, even if the sounds weren’t words yet? Those are the early indicators that the input is landing. Our SLP told me to think in eight-week cycles. If after eight weeks of consistent daily practice you’re seeing no change in any form of communication, that’s a signal to revisit the plan with your therapist. It doesn’t mean you failed. It means the approach might need adjusting.
Can grandparents or other caregivers do this too?
Absolutely, and they should. The more adults in the child’s life who use these strategies, the more practice opportunities the child gets. I typed up a one-page cheat sheet for my mom: offer two choices and wait, model words instead of quizzing, narrate what you’re doing, celebrate any attempt. She keeps it on her fridge. The key instruction I gave her was “don’t ask him to say it. Just say it yourself and wait.” That single shift, from testing to modeling, is the hardest habit for most adults to adopt, but once grandparents and babysitters understand it, they become part of the therapy team without even realizing it.
