AAC device definition: what it is, who needs one, and how it works

Last updated 2026-07-09

What counts as an AAC device

An AAC device is any tool or strategy that adds to or stands in for spoken language when someone can't rely on speech alone to get a message across. "Augmentative" means the tool builds on whatever speech a person already has. "Alternative" means it fills in when speech isn't available at all. Most people who use AAC are actually augmentative users: they have some speech, but need a backup for the moments when words don't come, whether that's from stress, tiredness, or too much going on around them.

ASHA defines AAC as "all forms of communication (other than oral speech) that are used to express thoughts, needs, wants, and ideas" ^[1]. That's a wide net on purpose. A sticky note a child points to counts. So does a speech-generating tablet that costs thousands of dollars.

Parents often picture expensive technology the moment they hear "AAC device," but that's really just one piece of it. Professionals split AAC into two groups: unaided and aided. Unaided AAC needs no outside tools at all, and covers sign language, gestures, and facial expressions. Aided AAC brings in something external, anything from a simple picture card to a speech-generating device (SGD).

When people say "AAC device" in everyday conversation, they almost always mean an aided system, usually one that produces speech. Still, it helps to know the unaided category exists, since plenty of children use both together, and that pairing often works better than either one alone ^[2].

The four main types of AAC systems

Professionals sort aided AAC into four categories based on how much technology is involved and whether the device talks. Knowing the categories helps you ask sharper questions at an evaluation.

No-tech and low-tech systems are usually where teams start with young children. The Picture Exchange Communication System (PECS) is probably the most researched low-tech approach: a 2006 randomized trial in the Journal of Autism and Developmental Disorders found meaningful gains in spontaneous communication after PECS training ^[3]. Low-tech doesn't mean less serious, either. Plenty of kids use paper-based PODD (Pragmatic Organisation Dynamic Display) books their whole lives, and those books carry vocabulary that goes far deeper than people expect.

Mid-tech devices like single-message buttons are cheap and surprisingly powerful for young children who just need one reliable way to say "I want more" or "help me." A speech-language pathologist can record a parent's voice into these buttons, and some kids respond to that far better than to synthesized speech.

High-tech dedicated speech-generating devices are purpose-built hardware: durable, often waterproof, with deep vocabulary systems already built in. Insurance, including Medicaid, tends to cover these more reliably than app-based systems, since they qualify as durable medical equipment under Medicare guidelines ^[4]. The tradeoff is cost and the learning curve families face getting used to them.

App-based systems on consumer tablets cost less up front and draw less attention socially (a kid holding an iPad doesn't get stared at the way a dedicated device sometimes does). But they trade away durability and insurance coverage. A dropped iPad with a cracked screen at the grocery store is a real crisis for a nonspeaking child who depends on it to talk. That's why many families end up running both: a dedicated SGD for school, and an app on a tablet as backup.

AAC isn't just for nonspeaking children

This myth has done real damage in the field. AAC is for anyone whose speech doesn't fully cover their communication needs in every situation, including kids who speak some words but lose access to them under stress or fatigue, something that shows up often in autism and in childhood apraxia of speech. It includes kids who are easy to understand at home but unintelligible to strangers, and adults recovering from a stroke who still have the language in their heads but can't get speech out reliably.

ASHA is direct about this: AAC is appropriate across a wide range of conditions, including autism spectrum disorder, apraxia of speech, cerebral palsy, Down syndrome, ALS, and traumatic brain injury ^[1].

A lot of parents worry that giving a child AAC will make them stop trying to talk. It's an understandable fear, but the research doesn't back it up. A 2006 systematic review in the American Journal of Speech-Language Pathology found that AAC intervention didn't suppress speech in any of the studies it looked at, and many participants actually spoke more after AAC was introduced ^[5]. That result has held up consistently over the two decades since.

Once you sit with the logic, it makes sense. Communication that keeps failing is exhausting. A child who has a reliable way to get a message across ends up less frustrated, more willing to communicate in general, and has more mental space left to keep experimenting with speech.

Conditions that often lead to an AAC evaluation

Anything that affects the motor planning, language processing, or physical production of speech can be a reason to look into AAC. A few diagnoses come up again and again in children.

Autism spectrum disorder is one of the most common, particularly when spoken language is limited or absent. Estimates vary, but roughly 25 to 30 percent of autistic individuals are minimally verbal or nonspeaking, depending on how those terms are defined ^[6]. Speech therapy for autism now tends to bring in AAC from the start instead of waiting to see how speech develops on its own.

Childhood apraxia of speech (CAS) is another. It's a motor speech disorder where the brain struggles to plan and program the movements speech requires, and many kids with CAS use AAC as a bridge while intensive speech therapy continues alongside it.

Down syndrome is a bit different: most children with Down syndrome develop some functional speech, but intelligibility is often a real challenge, so AAC tends to supplement speech rather than replace it for most of these kids.

Cerebral palsy that affects speech motor control, Rett syndrome, and acquired conditions like traumatic brain injury or pediatric stroke round out the list of common reasons families end up exploring AAC.

Late talkers, meaning children under 3 with delayed but not absent expressive language, are sometimes introduced to light-tech AAC as part of early intervention services. That doesn't mean anyone expects the child to stay nonspeaking; it's simply a practical way to support communication while speech is still catching up.

If you're unsure whether your child's situation calls for an evaluation, the ASHA resource page for families is a reasonable place to start ^[1]. But whether your child needs AAC is a question for an SLP with specific AAC training to answer, not an app or a checklist.

How does a child qualify for an AAC device?

There's no minimum age and no IQ cutoff. Under IDEA (the Individuals with Disabilities Education Act), schools are required to consider AAC as assistive technology for students who qualify ^[7]. Medicaid and most private insurers will cover a dedicated speech-generating device as durable medical equipment once a qualified professional documents medical necessity.

Getting a dedicated device usually follows the same route. An SLP with AAC training does a feature-matching evaluation first, looking at motor skills, vision, how much language the child understands, and what they need to communicate across different settings. That SLP then writes a letter of medical necessity explaining why a particular device or system fits the child, and from there the family submits paperwork to insurance or applies through the school district, depending on the path. Medicare and Medicaid require that speech-generating devices be prescribed for a documented medical need and that the device will actually be used, per CMS guidelines ^[4].

A good evaluation involves trying out more than one system. If someone recommends a single product without letting the family trial alternatives first, treat that as a warning sign.

Families who can't get in to see an SLP right away still have options. AAC lending libraries and state assistive technology programs often loan out devices, and the AT3 network (National Assistive Technology Act Programs), funded under the Assistive Technology Act, runs in every state offering device trials and training free or at low cost ^[8].

For school-age kids with an IEP, the district can provide an AAC device and training at no cost to the family under IDEA. That device technically belongs to the school, but the team still has to make sure the child can use it to communicate everywhere, not just during school hours ^[7].

How much does an AAC device cost?

It depends entirely on which kind of system you mean. Prices run from free to well over $10,000.

Low-tech options, things like printed boards, PECS card sets, or homemade communication books, cost almost nothing if you build them yourself with free symbol libraries and open-source tools. Commercial kits run $50 to $200. Mid-tech single-message buttons from vendors like AbleNet land in the $30 to $150 range per device.

High-tech dedicated speech-generating devices from companies like Tobii Dynavox or PRC-Saltillo are the big expense, typically $3,000 to $10,000 before insurance. Medicare and Medicaid will cover these as durable medical equipment when a family has a qualifying diagnosis and a letter of medical necessity ^[4]. Private insurance is far less predictable and varies a lot by plan.

App-based systems on a regular iPad cost less up front: $150 to $350 for the app itself, plus whatever the tablet costs. Insurance almost never reimburses this route, since an iPad isn't classified as durable medical equipment on its own. There are exceptions: a handful of insurers will cover an iPad if it's bundled into a medically necessary AAC system, but that takes specific documentation and isn't something to count on.

Waiting on an insurance decision doesn't have to mean waiting to communicate. Some SGD manufacturers run rental or loaner programs while claims are processed, and some states have Medicaid waiver programs set aside specifically for assistive technology. The AT3 network is a genuinely useful place to start looking, not just a name on a list ^[8].

So what does a family actually end up paying? If insurance approves a dedicated SGD, it can be close to $0 out of pocket. The iPad-and-app route usually lands somewhere between $500 and $1,000 total, before factoring in what it costs to replace the device if it breaks.

What is a speech-generating device, and how is it different from a basic AAC tool?

A speech-generating device (SGD) produces electronic voice output, either pre-recorded (a parent or child records phrases ahead of time) or synthesized through text-to-speech. It's a specific category within the broader world of AAC devices, not a synonym for it.

Every SGD counts as an AAC device, but plenty of AAC devices aren't SGDs. A picture exchange card is a good example: it's AAC, but since it produces no voice on its own, it doesn't qualify as an SGD.

That distinction shows up in two places day to day. Insurance and Medicare treat SGDs as their own DME category with defined coverage rules, while low-tech AAC tools generally aren't reimbursable at all ^[4]. Voice output also brings real communication advantages beyond paperwork: a message can carry across a room, the listener doesn't have to hover over the device to read it, and many kids and adults find it more motivating to use, since it sounds like actual speech.

Today's SGDs come in three basic forms. Dedicated hardware devices are purpose-built and ruggedized, usually the most reliable choice over the long haul. App-based systems run on regular tablets, mainly iPad, and cost less while offering more flexibility. Eye-gaze systems use cameras to track eye movement, letting someone with severe physical limitations select symbols just by looking at them, which opens up AAC for people with conditions like ALS or high-level spinal cord injury.

The vocabulary loaded onto the device matters just as much as the hardware. Core word systems (LAMP, SNAP, Unity, Proloquo2Go, among others) build language around the roughly 200 words that make up about 80 percent of what people actually say day to day, with fringe vocabulary such as specific names or places layered in alongside. Done well, a system like this lets a child build new sentences instead of just requesting the same handful of items over and over.

For a wider view of where this fits in, it's worth reading up on how speech therapy and speech-language pathologists factor into all of this.

Does AAC actually work?

Yes, though what counts as "working" needs some unpacking.

AAC research has grown a lot over the last 20 years, and the strongest evidence is for high-tech speech-generating devices combined with naturalistic developmental behavioral intervention approaches in children with autism. A 2018 meta-analysis in the Journal of Autism and Developmental Disorders reviewed 23 studies and found that SGD use was linked to real improvements in how often children communicated and how many words they used ^[9].

Among low-tech approaches, PECS (Picture Exchange Communication System) has the longest randomized trial record. Magiati and Howlin's 2006 RCT found that children trained on PECS made greater gains in initiating communication than a comparison group, though the size of those gains varied quite a bit from child to child ^[3].

For children with childhood apraxia of speech, pairing AAC with intensive speech therapy (think DTTC or Nuffield) tends to beat either approach alone. The research specific to CAS is thinner than for autism, but clinicians are confident enough in it that ASHA includes CAS in its AAC practice portal ^[1].

Nobody has great long-term data comparing outcomes by device type, honestly, because new devices keep showing up faster than researchers can run trials on them. What the field agrees on is simpler anyway: AAC works when there's enough vocabulary loaded onto it, when the adults around the child, parents, teachers, speech-language pathologists, are trained to model its use, and when the system goes wherever the child goes. A device sitting in a backpack during dinner isn't a communication system. It's an expensive paperweight.

Aided language input, where an adult points to or activates the AAC device while talking to the child, shows up again and again in the research as one of the strongest predictors of whether a child ends up using their device ^[2]. Parents modeling on the device isn't a nice extra. It's the intervention itself.

How is AAC different from sign language?

Sign language actually counts as AAC, in what's called the "unaided" category. American Sign Language and other signed languages aren't English on the hands: they're complete, fully grammatical natural languages with their own syntax.

Key word signing and Makaton (a simplified signing system used as AAC) work differently from ASL. Instead of standing alone as a language, they use a smaller set of signs alongside speech, following spoken word order. Many SLPs teach key word signing to young children because the motor control needed for hand movements often develops before the oral motor control needed for speech does.

For most families the real question isn't sign versus device, it's which combination of strategies covers their child across different environments and people. Most AAC users mix modalities rather than picking one: a child might sign "more" at the dinner table, point to a picture card at school, and use a speech-generating app somewhere loud. That kind of layering is normal, and it's what most SLPs recommend^[1].

One real difference is worth naming: signing only works if the person on the other end understands signs, while a speech-generating device doesn't need that. For families with relatives who don't sign, or for everyday errands out in the community, a device with voice output can do things signing alone can't.

If you suspect your child might need AAC, start with an evaluation from a speech-language pathologist who actually has AAC experience, since not every SLP does. When you call a clinic or school district, ask directly whether their SLP has experience conducting AAC feature-matching evaluations. That phrase signals you know what you're asking about, and it tends to get you a straighter answer. If your child is under 3, contact your state's early intervention program. Under IDEA Part C, children birth to 3 with developmental delays qualify for evaluations and services at no cost to the family^[7]. Early intervention teams often include SLPs and assistive technology specialists who can bring AAC into a wider communication plan. If your child is school age and already has an IEP, request an assistive technology evaluation in writing. Schools are required to consider AT, which includes AAC, as a related service, so put the request in writing and keep a copy for your records.

You don't have to wait for professional support to start helping at home. Print a core word board (there are dozens of free templates online), point to pictures as you talk, and look into aided language input. None of it costs money or requires an appointment. Some families also find AI-supported tools useful for building daily practice between sessions. Little Words, for instance, is an AI speech companion designed for neurodivergent kids, and it can support practice at home alongside professional AAC services, though it won't replace an SLP-led evaluation. For a wider look at what speech therapy involves for autistic children, from evaluation through intervention, see the autism spectrum speech therapy overview.

AAC isn't just for kids, either. Adults end up needing it after a stroke, after a diagnosis like ALS or Parkinson's disease, following a traumatic brain injury, or as part of a progressive neurological condition, and there's a solid body of research behind AAC use in adults, especially within the ALS population. Medicare Part B covers dedicated speech-generating devices as durable medical equipment when certain diagnostic and medical necessity criteria are met^[4], which in practice means documentation showing the person has a severe expressive speech disorder and genuinely needs the device.

Some adults with autism grew up nonspeaking without ever getting access to AAC, and more speech therapy practices now offer AAC evaluations built for adults in that situation. There's no age where the potential benefit stops. Our piece on speech therapy for adults covers what evaluation and services usually look like. The number of AAC users sharing their lives publicly, through blogs, videos, and advocacy work, has grown quite a bit, and groups like Communication First (communicationfirst.org) represent AAC users and push for their communication rights. Following AAC users directly is probably the best education a parent of a newly diagnosed child can find, since it shows what communication can look like across a whole lifetime. People sometimes use "AAC" and "PECS" as if they mean the same thing, but they don't. AAC is the broad category: any tool or system that supports communication beyond spoken words. PECS (Picture Exchange Communication System) is just one specific low-tech method within it. Andrew Bondy and Lori Frost developed PECS in the late 1980s for children with autism who weren't using speech in a functional way. The system teaches a child to physically hand a picture card to another person in exchange for something they want, and that hand-over is really the whole point: the child learns that initiating communication gets them what they're after.

PECS follows a defined six-phase protocol, and families and teachers are trained to work through it step by step. The research holds up reasonably well for Phases I through IV, which cover initiating requests, but it's weaker for Phases V and VI, which involve answering questions and commenting. It was never meant to be a full, lifelong communication system on its own, and plenty of children start with PECS and then move on to a higher-tech speech-generating device once their needs outgrow what picture cards can offer.

Other low-tech AAC options include core word boards, PODD (Pragmatic Organisation Dynamic Display) books, and LAMP (Language Acquisition through Motor Planning), which is really a vocabulary and teaching approach that can run on different devices rather than a physical system like PECS. If a school or clinic tells you "we use PECS" as though that's the entire AAC plan, it's worth asking more questions. PECS works well for early requesting, but it isn't a vocabulary system built to carry a child's full language development, and for most kids over age 4 or 5, a solid AAC program needs to look beyond it^[1].

AAC isn't a separate track from speech and language work, either; it's part of the same picture. Speech-language pathologists who introduce an AAC device aren't giving up on spoken language. They're looking at what a child understands, what a child can express, how a child uses language socially, and how all of that connects to reading and writing, with the device serving as one tool inside that larger plan.

Something that catches families off guard: AAC can actually support literacy. Many AAC vocabulary systems pair pictures with printed words, which helps kids connect symbols to the words they represent, and for children learning to read, that overlap between AAC use and early literacy tends to work in their favor^[2]. Echolalia, when a child repeats phrases they've heard before, shows up often in autistic children, and it doesn't run counter to AAC use at all; it's a form of communication in its own right. Reading up on echolalia and what echolalia actually looks like can help parents notice what their child is already communicating before any formal AAC system enters the picture.

An SLP's real job is to figure out where a child currently stands in communication development, then build the right mix of approaches, whether that's specific therapy techniques, an AAC tool, language modeling, or changes to the child's environment, to move them forward from there. AAC just happens to be one of the more flexible pieces of that toolkit. If your child has been identified as a late talker and you're trying to figure out where AAC might fit, early intervention services are the right place to start: getting support earlier consistently leads to better outcomes across communication disorders, something both ASHA and the AAP back in their clinical guidance^[1][10].

For families who want to keep building between sessions, Little Words offers AI-supported speech practice built for neurodivergent kids, and a short quiz can help you see what might fit your child's profile.

What parents usually want to know about AAC

AAC stands for Augmentative and Alternative Communication. "Augmentative" means adding to speech a child already has; "alternative" means stepping in when speech isn't available at all. ASHA defines it as any form of communication other than oral speech used to express thoughts, needs, wants, and ideas, so the category covers a lot of ground: gesture, sign language, and high-tech speech-generating devices all count.

There's no minimum age to start one. Researchers and clinicians have introduced simple picture symbols or basic gestures to infants as young as 9 to 12 months, on the theory that a child shouldn't have to wait for a way to communicate. Children from birth to age 3 qualify for early intervention evaluations under IDEA Part C, and AAC can be built right into that plan.

One worry comes up again and again: will AAC stop a child from talking? The research says no. A systematic review in the American Journal of Speech-Language Pathology found that AAC intervention didn't suppress speech in any study reviewed, and many kids actually spoke more once AAC entered the picture. Give a child a reliable way to get a message across and frustration drops, which tends to raise their overall drive to communicate, speech included.

Coverage depends heavily on the type of device. Dedicated speech-generating devices count as durable medical equipment under Medicare and Medicaid once a physician or SLP documents medical necessity, though private insurance varies plan to plan. App-based AAC on a regular tablet is rarely reimbursed, since the tablet itself doesn't qualify as durable medical equipment. School-age kids with an IEP can often get a device through their district at no cost under IDEA.

A speech-generating device (SGD) produces electronic voice output, either synthesized text-to-speech or recorded messages. Tobii Dynavox and PRC-Saltillo devices, along with iPad apps like Proloquo2Go, are common examples, and an SGD is usually what people picture when they hear "AAC device." Some are a single button with one message; others hold thousands of vocabulary items, accessed by touch, switch, or eye gaze. An AAC evaluation is run by a speech-language pathologist with specific training in this area, ideally alongside a team that might include an occupational therapist for motor access questions, a teacher, and the family. The process looks at motor skills, language comprehension, how the child currently communicates, and what they need day to day, then tries out a few systems before landing on a recommendation.

PECS, the Picture Exchange Communication System, isn't another name for AAC as a whole, it's one specific low-tech approach that teaches children to hand picture cards to a partner to make requests. It has decent evidence behind it for building early requesting skills, though most children who start with PECS eventually need a broader vocabulary system as they grow.

There's no cutoff on the other end of life either. Nonspeaking autistic adults can get AAC evaluations through speech therapy practices, and dedicated SGDs are covered under Medicare Part B as durable medical equipment when documentation shows a severe expressive speech disorder. Plenty of autistic adults who grew up without access to AAC have since found systems that work well for them. It really is never too late.

Core vocabulary refers to a small set of words, roughly 200 to 400, that make up 75 to 80 percent of what people actually say day to day: words like "more," "go," "want," "stop," "help," and "I." Most current AAC systems build around this core because it lets a child construct flexible sentences rather than just naming items, with fringe vocabulary (the specific nouns and names) filling in around it.

Aided language input, sometimes called modeling, means a communication partner points to or activates words on the AAC device while talking naturally with the child, similar to how kids pick up spoken language through immersion. Research keeps pointing to this as one of the strongest predictors of how much a child ends up using AAC and how their language grows. Parents are meant to model with the device themselves, not just prompt the child to use it.

An iPad is a general-purpose consumer tablet; an AAC device is anything built specifically for communication. The two overlap when someone loads an AAC app onto an iPad, but dedicated AAC hardware is purpose-built, with tougher cases, specialized mounting options, and longer warranties. Insurance and Medicare classify dedicated devices as durable medical equipment, while a plain iPad generally isn't reimbursable on its own.

Unaided AAC needs no outside tools at all: sign language, gestures, and facial expressions all count. Aided AAC involves something external, anything from a printed picture board to a high-tech device with eye-gaze access. Most AAC users end up combining both, and most SLPs recommend a multimodal approach rather than locking into just one type.

If you're trying to decide whether it's time for an evaluation or just more waiting, the American Academy of Pediatrics recommends a speech-language evaluation for any child not using words by 12 months, using fewer than 50 words by 24 months, or not combining two words by 24 months. If a child has a diagnosed condition that affects speech, such as autism, apraxia, Down syndrome, or cerebral palsy, it's worth requesting an AAC evaluation alongside speech therapy early on rather than waiting, since support started earlier tends to lead to better outcomes.

Families who can't afford a device yet still have options. The AT3 network, funded under the Assistive Technology Act, operates in every U.S. state and offers device loans, trials, and training at low or no cost. Many states also run Medicaid waiver programs for assistive technology, free core word boards and symbol libraries are easy to find online, and some SGD manufacturers will loan a device while an insurance claim works its way through.

Sources

1. ASHA, Augmentative and Alternative Communication Practice Portal: ASHA defines AAC as all forms of communication other than oral speech used to express thoughts, needs, wants, and ideas, and lists qualifying conditions including autism, apraxia, and cerebral palsy
2. Romski M & Sevcik RA, Augmentative Communication and Early Intervention: Myths and Realities, Infants and Young Children, 2005: AAC does not hinder speech development; multimodal AAC and aided language input are associated with improved communication outcomes in young children
3. Magiati I & Howlin P, Journal of Autism and Developmental Disorders, 2003: PECS-trained children showed greater increases in spontaneous communication initiations compared to control group in randomized comparison
4. CMS, Medicare Coverage of Speech-Generating Devices, Centers for Medicare and Medicaid Services: SGDs qualify as durable medical equipment under Medicare Part B when prescribed for a patient's medical need and documented by a physician or SLP
5. Millar DC, Light JC, Schlosser RW, American Journal of Speech-Language Pathology, 2006: Systematic review found AAC intervention did not suppress speech in any study reviewed; many participants showed increases in speech after AAC introduction
6. Tager-Flusberg H & Kasari C, JAMA Pediatrics, 2013: Approximately 25 to 30 percent of autistic individuals are minimally verbal; researchers note heterogeneity in how minimally verbal is defined across studies
7. U.S. Department of Education, Individuals with Disabilities Education Act (IDEA), 34 CFR Part 300: IDEA requires schools to consider assistive technology, including AAC, for eligible students; Part C covers birth to age 3 at no cost to families
8. AT3 Center, Assistive Technology Act Programs: The AT3 network, funded under the Assistive Technology Act, operates in every U.S. state and provides device demonstrations, loans, and training at low or no cost
9. Alzrayer N, Banda DR, Koul RK, Journal of Autism and Developmental Disorders, 2017: Meta-analysis found SGD use associated with significant improvements in communication frequency and vocabulary size in children with autism across 23 reviewed studies
10. American Academy of Pediatrics, Policy Statement on Early Intervention: AAP recommends developmental surveillance and early intervention referral for children not meeting speech-language milestones, noting earlier support produces better outcomes