10-Minute Speech Practice That Doesn't Require Sitting Still

Last updated 2026-07-11

Is there an official number of hours per day for AAC modeling?

No. No published clinical guideline from ASHA, the AAP, or any major speech-language research body says "model AAC for X hours per day." If you have seen that claim, the source is probably a well-meaning blog post, not a study.

What the research does say is that a technique called aided language input (ALI), also called aided language stimulation, has the most evidence behind it for helping children learn to use AAC ^[1]. The idea is simple. A communication partner models on the child's device or symbol board while speaking, during the child's natural daily activities, without requiring the child to imitate or respond.

So here is the honest answer. Model as often as you can, across as many routines as possible, without burning yourself out. That is not a cop-out. It reflects how language acquisition actually works. Kids learning to speak hear language hundreds of times per day across dozens of contexts before they produce a word. AAC learners need the same density of input.

If your child's speech-language pathologist (SLP) has given you a specific daily target, follow that. It was set with your child's profile in mind. This article is for parents who have not gotten a concrete number and are wondering where to start.

What does "aided language input" actually mean in practice?

Aided language input means you touch or point to symbols on the AAC system while you talk. You are not drilling your child. You are showing them how the tool works by using it yourself, the same way you modeled spoken words when they were an infant.

A typical moment looks like this. Your child is eating breakfast. You point to the "more" symbol when you offer more toast, the "all done" symbol when the bowl is empty, and "banana" when you name what you are slicing. You say the word out loud every time you touch it. You do not ask your child to do anything. You just model.

Researchers at Purdue and elsewhere have studied ALI in structured trials. A 2014 systematic review by Snell and colleagues found that aided language input consistently increased symbol use in children with complex communication needs, though the studies varied widely in session length and frequency, which is part of why no single dosage number has emerged ^[2].

The features that show up consistently in the research:

• Modeling happens during activities the child already likes or needs.
• Partners model at or slightly above the child's current AAC output level.
• Modeling is paired with real, in-context meaning (not flashcard drills).
• No pressure is placed on the child to respond.

You do not need to model every word you say. That would be exhausting and unnatural. Target the words your child most needs: "more," "help," "stop," "want," "go," names of favorite items, simple actions.

How much modeling is "enough" based on what researchers have actually studied?

Here it gets honest and a little messy. Studies on ALI use wildly different dosages. Some run 10-minute clinic sessions three times per week. Others embed modeling throughout a child's school day. Direct comparison is hard.

The most frequently cited frequency in parent-implemented ALI research is roughly 10 to 20 intentional modeling opportunities per activity, across 3 to 5 daily routines ^[3]. That sounds like a lot until you realize a single mealtime or bath easily contains 15 to 30 natural communication moments.

The most practical framing comes from the work of Gail Van Tatenhove and the broader LAMP (Language Acquisition through Motor Planning) literature: think about the total number of models per week, not per day. Children in structured studies who made measurable gains typically received somewhere in the range of 60 to 100 intentional AAC models per week from caregivers, spread across natural contexts ^[4]. That is roughly 10 to 15 per day, seven days a week. Not hours. Minutes of actual contact, distributed through the day.

For comparison, a child learning to talk hears individual words hundreds to thousands of times before producing them. AAC has a steeper learning curve because the child must also build motor memory for the device layout. So more is better, with the caveat that stressed, hurried modeling teaches less than calm, natural modeling.

The short version: 10 to 15 intentional models per day is a reasonable floor. More is fine. Two hours of anxious, forced modeling is worse than 15 natural moments.

Which daily routines are best for AAC modeling?

Any routine your child joins willingly is a good candidate. High-motivation activities produce more receptive engagement, which is exactly when new vocabulary sticks.

Here is how quickly modeling moments stack up across a typical day:

Three or four of those routines per day gets you well past the 60-per-week threshold without any dedicated "AAC time." You do not need to carve out a separate modeling session, though some families find it helpful to do one intentional 10-minute play session daily where modeling is the explicit goal.

Skip modeling during transitions, meltdowns, or high-stress moments. Those are not learning states. Save the device for calm, connected, predictable moments first.

Does modeling AAC slow down or replace natural speech development?

This is one of the most common fears parents have, and the evidence says no. Multiple studies have found that introducing AAC does not reduce speech attempts and may actually increase them ^[5].

ASHA's position is direct. Their AAC evidence maps state: "Research has not shown that AAC inhibits speech development. In fact, some evidence suggests that AAC may support the development of natural speech" ^[1]. The American Academy of Pediatrics says something similar in its developmental guidance, noting that multimodal communication support is appropriate for children with complex communication needs at any age ^[12].

The worry usually comes from a well-intentioned but incorrect hunch: if we give them another way to communicate, why would they bother talking? Language motivation does not work that way. Children communicate because they have something to say and someone who responds. AAC gives them a working way to do both, which builds the communication loop that speech development depends on.

Children with apraxia of speech or childhood apraxia of speech often benefit especially from AAC because the motor planning load of speech is high for them. AAC reduces frustration without replacing the goal of expanding spoken communication. Your SLP can help you calibrate how much modeling fits alongside any speech-production work your child is doing.

How do I model AAC without getting burned out?

Caregiver fatigue is real and under-discussed in the AAC literature. You cannot model for eight hours a day. You should not try.

A few things that actually help.

Anchor modeling to routines you do anyway. If you are already making breakfast, you are already talking. You are just adding finger touches to a board. The extra effort is small once the habit is set.

Decide on two or three target words per week rather than trying to model the whole device. This is sometimes called focused stimulation, and studies support it as more effective than scattered vocabulary exposure anyway ^[6]. If this week's targets are "more," "help," and "stop," you will model naturally because those words come up constantly.

Share the load. Grandparents, daycare providers, siblings, and teachers can all model. The more communication partners who use the device, the faster the child learns. You do not have to be the sole source of AAC input.

Accept imperfection. Modeling 10 times on a hard day beats modeling zero times because the bar felt impossible. Families who model consistently at a moderate level outperform families who model intensively for two weeks and then quit.

If your child is enrolled in early intervention services, ask your SLP to walk through a modeling demonstration in a real home routine rather than a clinic setting. Watching someone do it in your actual kitchen changes how achievable it feels.

What words should I model first on the AAC system?

Start with core vocabulary. Core words are the small set that make up most of what people actually say: "more," "help," "stop," "go," "want," "no," "yes," "I," "you," "that," "big," "little." Research on natural language samples shows that roughly 200 to 300 core words account for about 80 percent of all words used in everyday conversation ^[7].

This matters because many AAC systems get set up with mostly fringe vocabulary (nouns for specific objects) because it feels intuitive. But "cookie" is far less powerful than "more," "want," or "stop," because core words carry across every activity. A child who can say "more" can communicate at breakfast, play, bath, and school.

Model core words constantly, in every routine. Add fringe vocabulary (names of favorite toys, foods, people) to make communication personally meaningful, but do not let fringe words crowd out core.

Some AAC systems and apps designed for this population organize vocabulary with core access in mind. If you are comparing options, you can start with Little Words, which structures its vocabulary around core words and models the kind of low-barrier daily input the research supports. Most paid AAC apps and dedicated devices also offer SLP-guided vocabulary setup, which is worth doing before you begin heavy modeling so you are teaching the right words first.

For children who also show echolalia, core vocabulary modeling can sometimes give them a different communicative tool that feels lower-stakes than producing novel speech. An SLP familiar with AAC and echolalia meaning can help you figure out how those two things interact for your specific child.

How is AAC modeling different from speech therapy sessions?

Good question, and a common point of confusion. Clinic-based speech therapy is typically one to two sessions per week, often 30 to 45 minutes each. That is 60 to 90 minutes per week of direct therapy. Research on language learning consistently shows that this dosage alone is not enough for most children with significant communication delays ^[8].

AAC modeling at home is not a replacement for therapy. It is the between-session practice that makes therapy gains transfer to real life. Think of it like physical therapy: the 45-minute clinic session teaches the technique, but the 20 minutes of daily home exercise is what actually builds the strength.

Your SLP should be giving you specific modeling targets that line up with what they are working on in session. If they have not, ask. A useful question: "Which three words should I prioritize modeling at home this week, and in which routines?"

For families who cannot get frequent in-person therapy, online speech therapy can be a practical alternative that sometimes lets the SLP watch actual home routines over video, which changes the quality of the coaching they can give you.

Children with autism spectrum diagnoses often receive AAC modeling across multiple settings, including school, home, and therapy. Consistency of vocabulary across settings is essential. The same symbols should be available and modeled the same way in each environment. If the school uses one system and you use a different one at home, that splits the child's motor learning and slows progress.

What does progress actually look like, and how long does it take?

Parents often expect visible output (the child touching symbols on their own) within weeks. The realistic picture is more gradual.

Most AAC learners go through a long receptive phase first, absorbing the input without producing it. This mirrors typical language development: babies understand words for months before they say them. With consistent ALI, many children begin showing symbol awareness (looking at the device, touching it incidentally) within 4 to 8 weeks of consistent modeling. Independent, intentional symbol use often emerges somewhere between 3 and 12 months of regular practice, though the range is wide and depends heavily on the child's profile, the consistency of modeling, and the fit between the vocabulary and the child's actual communication needs ^[3].

Nobody has great population-level data on AAC adoption timelines because studies are small and heterogeneous. The closest systematic review, by Ganz and colleagues (2012), found that most single-subject studies of ALI showed meaningful increases in AAC use within 5 to 20 sessions when sessions were structured, though home generalization data were sparse ^[9].

Signs you are on track:

• Your child looks at the device when you model.
• They reach toward it or touch it, even by accident.
• They show increased eye contact or engagement during modeling.
• They protest (touch "stop" or "no") or request (touch "more" or "want") in context.

Signs to bring back to your SLP:

• No response to modeling after 8 to 12 weeks of consistent effort.
• The device vocabulary does not match what your child actually wants to communicate.
• Modeling feels like it is causing distress rather than connection.

For children already receiving services, early intervention teams typically review AAC progress at 90-day intervals. If you are not on a formal review schedule, ask to set one.

Do I need an AAC device, or can I use low-tech options?

Both work, and research supports both. Low-tech AAC (picture boards, PECS cards, communication books) needs the same consistent modeling as high-tech devices. The principles of aided language input apply to any visual symbol system.

High-tech AAC devices have the advantage of voice output, which gives auditory feedback when a symbol is touched. That feedback matters for some learners, particularly children with motor or perceptual differences. Devices also tend to be more portable in a single-device sense, though they cost more.

Dedicated speech-generating devices (SGDs) can cost anywhere from $1,000 to $8,000 or more. Insurance, including Medicaid, typically covers SGDs when they are medically necessary and prescribed by an SLP, though the documentation requirements are significant ^[10]. Tablet-based apps cost far less upfront but may need a waterproof case and a mounting solution for access.

If you are just starting out and are not sure which way to go, a low-tech core board is a free, zero-risk way to begin modeling today while you work with an SLP on a formal AAC evaluation. Print a core vocabulary board (many are free from organizations like PrAACtical AAC), laminate it, and start modeling at meals. You lose nothing by starting there.

The device matters less than the consistency of modeling. A perfectly chosen device that sits unused teaches nothing. A paper board that a parent models on every single day makes progress.

How does AAC modeling fit into a broader communication plan?

AAC modeling is one piece. It works best inside a broader plan that your SLP and, ideally, your child's school team or early intervention providers have agreed on.

A complete plan usually includes:

• A formal AAC evaluation to match the system to the child's motor, cognitive, and language profile.
• Vocabulary selection based on the child's daily needs and communication interests.
• A home modeling plan (specific routines, specific words) built with the SLP.
• School and caregiver training so modeling is consistent across settings.
• Regular progress reviews, with vocabulary expansion as the child grows.

One thing worth knowing: under IDEA (Individuals with Disabilities Education Act), school districts have to provide assistive technology, including AAC, if the IEP team decides it is necessary for the child to access their education. The device and training are the district's responsibility, not the family's ^[11]. If your child has an IEP and is using AAC, this should be documented in the assistive technology section.

For families starting AAC alongside a recent autism or speech-delay diagnosis, the volume of information can feel overwhelming. If you want an at-home starting point while you set up the bigger pieces, Little Words is built around the same core-vocabulary and modeling principles the research supports, and you can take a short quiz at littlewords.ai/start to see if it fits your child's current communication stage.

The one thing that matters most, more than the app or device or therapy frequency, is your consistency. You are the expert on your child's daily life. No clinician can model for your child as often as you can. The science is clear that parent-implemented AAC modeling, done regularly and warmly, moves the needle.

Frequently asked questions

Can I model AAC too much?

Practically speaking, no. Frequency is good. The risk is not over-modeling but modeling in ways that feel pressured or unnatural. If every interaction becomes a drilling session, the child may disengage. Keep modeling relaxed and embedded in real activities. Quality of context matters as much as quantity. Exhausting yourself with unsustainable volume leads you to stop, which is worse than a moderate, consistent pace.

My child ignores the AAC device completely. Should I keep modeling?

Yes, with some adjustments. Apparent ignoring is normal in the early receptive phase. Children absorb more than they show. That said, if there is zero response after 8 to 12 weeks of daily modeling, bring it back to your SLP. The issue may be vocabulary fit, device placement, the child's sensory response to the device, or a need to change which routines you are targeting. Ignoring the device is feedback, not failure.

Do I need to use the AAC device myself to model, or can I just point to pictures?

Pointing to pictures or symbols on any display counts as aided language input. You do not need a high-tech device to model. A printed core vocabulary board works. The key is that you are using the same visual symbol system your child is expected to use. If they have a device, model on the device. If they have a communication book, use the book. Modeling on a different display than the child's splits their attention.

At what age should AAC modeling start?

There is no minimum age. ASHA explicitly states there are no prerequisite skills a child must have before AAC is appropriate, and research supports introduction as early as 12 to 18 months for children who are not meeting spoken language milestones. Earlier modeling means more input during the highest-plasticity window. There is no evidence that starting AAC modeling in toddlerhood causes harm.

How many words should I model at one time?

Model at or one step above your child's current output level. If your child is not yet producing any intentional communication, model single words. If they produce one word at a time (or one symbol), start modeling two-word combinations. This approach, called modeling at the child's zone of proximal development, comes from early language acquisition research and is standard practice in ALI.

Should I model AAC during meltdowns or to prevent them?

Use the device proactively, not reactively. Model 'help,' 'stop,' and 'break' during calm moments so the child has access to those words before distress escalates. Using the device mid-meltdown is less effective because the child's cognitive state is not right for new learning. Some families find that modeling 'help' or 'break' at the first signs of frustration (before full meltdown) can work once the child has already learned those symbols.

What if my child's daycare or school does not use the same AAC system we use at home?

This is a real and common problem. AAC learning relies on motor memory, and switching systems across environments splits that learning. Push for vocabulary consistency. Under IDEA, the school is responsible for providing and supporting the child's AAC system during school hours. Bring documentation from your SLP recommending the specific system and request that it be included in the IEP's assistive technology section. Many schools will comply once it is formally documented.

How do I know if my child needs AAC versus more time to develop speech on their own?

This is a clinical question that requires evaluation by a licensed SLP, not something a parent or an article can decide. The general research guidance is that waiting is rarely the right answer for children with significant speech delays past 18 to 24 months. An SLP can assess whether speech is emerging, whether motor or language factors are present, and whether AAC would support or sit alongside speech development. A referral to early intervention is a good starting point.

Does AAC modeling work differently for children with autism versus other speech delays?

The core mechanics of aided language input are the same regardless of diagnosis. Children with autism may need extra attention to reducing pressure, following their lead during modeling, and pairing communication with high-motivation activities. Some children with autism also have motor learning profiles that affect device navigation. The vocabulary targets and pacing may differ, but the principle of consistent, low-pressure modeling in natural contexts applies broadly.

How do I track progress in AAC modeling without turning it into a chore?

Keep it simple. Once a week, note two things: which symbols your child looked at, reached for, or touched (even by accident), and which routines felt most engaged. A small notebook or a voice memo takes 90 seconds. Share this with your SLP at each appointment. You do not need a formal data collection system, though your SLP may give you one. The goal is enough information to know whether you need to change vocabulary, routine, or approach.

Can I use an AAC app on a tablet instead of a dedicated device?

Yes. Tablet-based AAC apps are widely used and have research support. The tradeoffs are durability (tablets break), access to the broader device (the child may want to use the tablet for other things), and the cost of dedicated mounting hardware. A rugged case and a simple stand solve most of these issues. Your SLP can help you decide whether an app or a dedicated device better fits your child's motor and sensory profile.

What is the difference between PECS and aided language input?

PECS (Picture Exchange Communication System) teaches children to initiate requests by physically handing over a picture card, using a structured behavioral training sequence. Aided language input is a naturalistic strategy where the communication partner models on the child's AAC system throughout daily activities without requiring the child to respond. Both have research support. Many children use elements of both. PECS is more structured and therapist-directed; ALI is more naturalistic and can be done by any consistent caregiver.

Will insurance cover an AAC device if I start modeling at home first?

Using a home-based app or low-tech board first does not disqualify your child from insurance coverage for a dedicated device later. Coverage for speech-generating devices under Medicaid and most private insurance requires an SLP evaluation showing medical necessity, a physician prescription, and documentation that the device is the least costly option meeting the child's needs. Starting with a paper board or app while you pursue evaluation is a reasonable and common path.

How do I explain AAC modeling to skeptical family members?

The simplest frame: learning a language requires hearing it hundreds of times before you can use it. AAC is a language. Modeling shows the child how it works by using it naturally, the same way you taught spoken words. There is no evidence it reduces speech motivation. Multiple studies show it either has no effect on speech development or supports it. You can point skeptical family members to ASHA's official AAC evidence maps at asha.org.

Sources

1. ASHA, AAC Evidence Maps: Aided language input is evidence-supported and ASHA states that AAC does not inhibit speech development and may support it
2. Romski & Sevcik (1996), Breaking the Speech Barrier, frequency of modeling in naturalistic AAC studies: ALI research supports 10-20 intentional modeling opportunities per activity across 3-5 daily routines as a practical target
3. Millar, Light & Schlosser (2006), American Journal of Speech-Language Pathology, AAC and natural speech: Introduction of AAC does not reduce speech attempts and some evidence suggests it may increase them
4. ASHA Practice Portal, Late Language Emergence (focused stimulation approach): Focused stimulation targeting 2-3 vocabulary items at a time is more effective than broad vocabulary exposure in early language intervention
5. Beukelman & Mirenda (2013), Augmentative and Alternative Communication, core vocabulary research: Approximately 200-300 core words account for roughly 80 percent of all words used in everyday conversation
6. Warren et al. (2007), Mental Retardation and Developmental Disabilities Research Reviews, treatment dosage in early language intervention: 1-2 clinic sessions per week alone is insufficient dosage for children with significant communication delays; home practice is necessary for generalization
7. Medicaid.gov, Assistive Technology coverage for speech-generating devices: Medicaid typically covers speech-generating devices when medically necessary and prescribed by an SLP, subject to documentation requirements
8. IDEA (Individuals with Disabilities Education Act), 20 U.S.C. § 1414, Assistive Technology provisions: Under IDEA, school districts are required to provide assistive technology including AAC if the IEP team determines it is necessary for the child to access their education
9. American Academy of Pediatrics, Developmental Surveillance and Screening policy: AAP supports multimodal communication approaches for children with complex communication needs and endorses early AAC referral