Josh's AAC App Design Research (Feedback Needed!)

[PullJosh]

Hi everyone! I’m Josh Pullen. Many of you are probably here because you saw me in a video where I discovered the need for a free AAC app in the world.

I am a developer, not an SLP or other AAC expert. My goal between now and the EdTech-a-thon event is to learn everything I can about these apps and make a plan for which features our AAC app should include.

I need your help. I’ve done my best to synthesize some online research and all the information that has been sent to me via DM from our videos. Based on that, I created a plan below for the features I think our app should have long-term. Now I need feedback from SLPs, parents, and other experts. Which parts of my notes are wrong? What am I missing? What do we need to consider? Please reply to this topic with anything and everything you think of.

App Plan

The EdTech-a-thon is a 3-day event, but building an AAC app is clearly a much longer, ongoing task. Our goal during the 3-day event is to build and ship a first prototype that is usable enough to test, and then continue to improve the app over the year(s).

I propose the following long-term design. I would love feedback on ALL of these points.

• Our app should be completely, 100%, absolutely free forever
• Our app should work on as many devices as possible.
  • It should work as an iOS app (phones & tablets)
  • It should work as an Android app (phones & tablets)
  • It should work in a web browser without installing anything (phones, tablets, computers)
• The grid
  • The board should be a rectangular grid, with tiles always remaining in the same physical location on screen (to help with motor planning)
    • The grid size (# of rows/columns) should be customizable
    • Feedback received: The app should include an onboarding wizard that helps parents choose the best grid size (because they often start with too few squares and then have to rearrange the icons later)
    • Feedback received: When using the same board across multiple devices such as phone/tablet, keep all the tile positioning the same and just stretch/shrink the grid.
  • Tiles
    • Each tile can have an icon and a text label (both are optional; see multi-lingual support details below)
    • When tapped, tiles can be configured to do one of a few things…
      • Add a word/phrase/sentence to the message bar (sometimes the same word/phrase as the label on the icon itself, but not always)
        • There should be a global, app-wide setting for “speak on selection” to also play the sound immediately on tap
        • Certain tiles can be configured to speak immediately without being added to the message bar
      • Navigate into a folder
      • Navigate home
      • Question: Should tiles be able to do anything else?
      • Feedback received: Press and hold to change grammar (I assume tense/plural/etc)
    • Tiles can be temporarily hidden/shown without being deleted
    • Tiles can be dragged & dropped into new locations on the board (without disrupting the rest of the layout)
    • Tiles have a background color. The app should make it easy to use a consistent color palette when customizing the board, and should provide a default set of colors aligned with the Fitzgerald Key
  • It should be easy to have multiple different boards installed & switch between them
  • Feedback received: Distinguish between core word boards and vocabulary boards.
  • Feedback received: Have a setting that can be turned on so that every time a vocabulary word is selected the user is automatically brought back to the core word screen (for easier motor planning)
  • We should support non-grid layouts and visual scenes, where buttons can be placed & sized anywhere on screen (not restricted to a grid) and a background image can optionally be added (for example, to allow users to tap on on the different play structures on a photo of a park)
  • We need a “find word” feature. When tapped, it should allow the user to search for a tile by name and then practice getting there by graying out everything other than the correct folder/icon to touch until reaching the word.
• Board Customization
  • It should be easy to customize existing boards
  • It should be possible to create completely new boards from scratch and share them with others
  • Users should be able to customize images/icons
    • By taking a photo using the camera
    • By uploading an image
    • By picking an icon from pre-made packs like Open Symbols
    • Question: Would a built-in drawing mode be useful? It would allow users to draw icons directly within the app, possibly with Apple Pencil support. Feedback received on this point has been very mixed so far.
    • Question: Would using emojis as symbols be a helpful option?
    • Question: Would AI icon generation be useful? (Type a word and it makes an icon for you)
    • Question: Are there any other image creation methods that would help?
  • Users should be able to customize audio
    • By using a built-in AI voice
      • Customizing the accent/gender/speed/etc is really important to help users feel comfortable with their voice. (I’m not sure how we can do this in a cost-effective way. Can we get ElevenLabs credits as a nonprofit?)
    • By recording using the built-in microphone
    • By uploading audio files (One AAC user messaged me explaining that they like to have sound effects and sound bites from movies available for fun communication)
  • Users should be able to share their customized boards
    • It should be possible to download the board as a file to be shared via email, Google Drive, Teachers Pay Teachers, AirDrop, etc.
      • Question: Is it helpful to support the open board format?
    • It should be easy to generate URLs or QR codes so that others can add your board to their device instantly
• Multi-lingual use
  • The entire app, including its UI, should be available in as many languages as possible
  • Users should be able to turn on multiple languages. When they do so, a language picker should be added to the UI that allows them to quickly switch the speaking language
  • Icons with text should display in the selected language
  • Question: Would it be helpful to let users set language-specific boards, or does switching languages just mean keeping the same board but in a different language?
  • Question: Are there other features we need for excellent multi-lingual support?
• Other Accessibility Needs
  • Scanning support
• Parental controls
  • We should enable setting a PIN that must be entered to modify the board/app settings
  • We should make it easy for users to turn on “guided access” (iOS) or “app pinning” (Android) to prevent closing the app. On Android, we can have a button within the app that instantly enables this. iOS is more limited; we can include instructions that teach users how to turn on guided access but we can’t make a button that does it automatically. (On the web, this might just mean going full-screen.)
• In-app training & help
  • Feedback received: In practice, families often struggle with supporting AAC-using children because parents/guardians aren’t familiar with these tools themselves and are learning alongside their child. Built-in help and support would dramatically increase the practical usefulness of the app.
  • Josh says: We should have built-in text and videos strategically placed throughout the app to help folks learn how to use the app and effectively support the AAC user. These support resources must be developed by SLPs, not app developers. I think we should focus on making the app work first, and then find SLPs who would be willing to help create the support resources later.

Bigger Questions

• I’ve seen a lot of talk about Gestalt vs Analytic language processing. I understand that gestalt processors start with full phrases and then break them down whereas analytic processors start with individual words and build them up. Are there features above and beyond the board customization I’ve explained above that are needed to support both use cases?
• There is clearly a TON of work that goes into designing board layouts, and that’s the world of SLPs, not app developers. As a first step this year, is it good enough for us to ship an app that is mostly an empty skeleton by default that allows users to build & share their own boards? I’m hoping that building boards could become a longer-term community effort, but it has to begin with building a functioning (albeit mostly empty) app. Does this approach feel effective? The feedback we’ve received is that it’s essential to ship with built-in boards, and we might be able to use Project Core as a starting point for picking a set of core words. I would still love more help/feedback around what our starting board should look like (while avoiding any legal/plagiarism issues!)
• I’ve seen talk about core words vs vocabulary, with the idea that core words should always remain a tap away and have a more stable layout, whereas niche vocabulary can be a bit more buried into an organized folder system and is more frequently customized. Aside from the board customization features described above, are there other features we need to specifically help handle core words vs vocabulary?
• We had a developer, AJ, reach out to us who is already building a free AAC app. AJ offered to let us use it as a starting point for free if we wish. If you have the time, we’d love feedback on that app as well.

We need your expert help!

• Please reply to this topic with any and all feedback about the app plan & questions above (feel free to highlight a specific part of the text above with your mouse and then click “quote” to reference that bullet point directly)
• I will continually update the app plan based on the feedback I receive
• Please check this forum periodically because we will be using it a ton during the actual event from July 20 - 22

From the bottom of our hearts, thank you so much for your time. We hope to make you proud and help the community!

A huge thank you to OpenAAC for their list of development considerations. Very helpful!

[myVoice]

• Yes, users should be able to customize images and icons. It would be helpful to offer an online icon search, such as through Google Images, as well as AI-generated icons. However, AI-generated icons may need refinement to ensure they look polished, consistent, and not obviously AI-generated.
• Adding topic-specific pages, similar to the Topics pages in TD Snap, would be extremely beneficial. These pages make it easier to access relevant vocabulary quickly during fast-paced, topic-based conversations.
• Audio should be highly customizable. Open-source voice options, such as those available through Piper, could provide users with greater flexibility and a wider range of voices.
• The ability to download and share customized boards is very important. Users should be able to share boards through email, Google Drive, AirDrop, Teachers Pay Teachers, and other platforms. Supporting the Open Board Format would be valuable for compatibility and portability. My only concern is that boards in this format are not always visually appealing by default. If the app could preserve the open format while presenting boards in a more polished and engaging way, that could significantly improve user and family buy-in.
• Yes—gestalt language processing should absolutely be considered. Topic pages could be particularly helpful because many gestalt language processors benefit from quick access to meaningful, situation-specific phrases. In my experience, these phrases often need to be personalized around the individual’s interests, routines, and preferred ways of communicating.
• I feel somewhat conflicted about launching with a mostly empty framework. Many families are initially excited about customizing an AAC system, but having too much to create at once can quickly become overwhelming and lead to burnout. I would recommend starting users with a strong core-word page, along with several prebuilt category pages containing essential and commonly used vocabulary. Families and professionals could then edit, personalize, and expand those pages over time. This would provide a functional starting point while still allowing the board library to grow through community contributions.
• Last thing: thinking about my job as an SLP, I see many patients via telehealth. As a result we rely a lot on the syncing features of TD snap, Dropbox for Grid 3, etc. I suggest trying your hardest to get it native to the app to where if someone customizes the app, they can sync it and it goes to the other apps (this relies on profiles that can be downloaded per aac user).
• Check out WeaveChat and their new app Context AAC for ideas!
• Last note: I am 1000% an SLP first but I have done quite a bit of coding with Claude, codex, and antigravity. I’ve tried making AAC software and any iteration I make tends to look like “AI slop” and I get burnt out with it. I’m very excited to see what someone with more coding knowledge is able to do!!

[oliviasloane12]

I’m a TA in a 6:1:5 classroom for nonverbal/high behavior students, here are some of my thoughts!

-There is so much value in having aac accessible on a watch, sometimes a child needs a visual and we don’t have aac nearby, having an index of visuals on my wrist would be a dream.

tiles- i do not think it is useful to for the device to speak without being added to the message bar. oftentimes hearing a word from the device from far away is difficult, and i would like to be able to walk up and see what the aac user has said when i was unable to hear.

tiles- make sure some core words are always available. TouchChat uses a format where the home page contains most core words, when you go to a certain folder commonly associated core words pop up (ex. in the vehicles folder the core words “go”, “on”, “fast”, “slow” and others are available) These folders are also where you will find more fringe vocabulary (ex. the car folder also has “garbage truck”, “excavator”, “train”, “airplane”)

voice- it is incredibly important to allow voice customization. I’m based in new york, one of my students had an australian accent coming out of her device. she also hated her device and would gag at it. after changing the color of the case, and making the voice sound more like her, engagement with the aac device increased.

voice- option to record your own audio should be included. family phrases, cultural dialects and other fringe things families may want to add might be difficult without this option. also my students love sound effects on their device that the SLP recorded.

[PullJosh]

Thanks so much! Every point here is very helpful.

[SNFSLP]

My responses are in bold with questions still included - intention was ease of navigating answers rather than let myself free flow everything. I 100% agree that the app should be free forever and work on as many devices as possible.

Question: Is there any value in supporting smart watches?

-I have not come across users that would benefit from smart watch use, but I work in the older adult/geriatric population. I would suggest that it not be a current focus, but potentially a future project as technology advances.

Question: What should happen if the same grid is used on a tablet and then on a phone? Just shrink it down but keep all the positioning the same? -Would it be possible to have an option for either “fit to size” or “wrap icons” based on user preferences? I’ve had some clients that thrive with consistent positioning and others that rely more on the ability to read each icon. Vision acuity plays a role, especially in the older population.

Certain tiles can be configured to speak immediately without being added to the message bar (Question: Is this a useful feature?) - I think that feature is useful. Sometimes, you want to get a quick word in without having to do multiple taps.

Question: Should tiles be able to do anything else? - Nothing is coming to mind right now.

It should be easy to have multiple different boards installed & switch between them - LOVE THIS

We should support non-grid layouts and visual scenes-YEEEEESSSS! Visual scenes are my favorite.

Question: Would a built-in drawing mode be useful? -I personally have not been is a situation where this would have been a helpful feature.

Question: Would using emojis as symbols be a helpful option? -I do love this idea.

Question: Would AI icon generation be useful? -I think it could be a cool feature, but if the icon creation is easy enough (like less than 2-3 minutes, I don’t think it would be necessary.

Question: Are there any other image creation methods that would help? -It would be nice if we could edit pictures we have taken for an icon. But I suggest that because I always upload a photo and then think “aw man, I need to crop that and brighten it up.”

Question: How important is it to allow customizing the gender/accent/etc of the voice? -HUGE! This is their voice. We should be letting them have control over how they sound. I have had clients who refused to even trial a device until I got the voice right.

By recording themselves speaking

Question: Is this actually useful, or is it weird if the speaking voice is different for different icons because it was recorded by different people? -I personally haven’t been in a situation where their own voice would be used. If it’s a scenario where the client has already completed word banking with their own voice, then definitely would be helpful. I’m hopeful that someone with experience working with word banking will comment/participate.

By uploading audio files (One AAC user messaged me explaining that they like to have sound effects and sound bites from movies available for fun communication) -This is excellent.

Question: Is it helpful to support the open board format? -I think it could be, but potentially difficult to manage.

It should be easy to generate URLs or QR codes so that others can add your board to their device instantly - Love this concept. It gives the person the opportunity to use their voice even if they left their device behind or the battery died.

I haven’t had much multi-lingual experience and fee unable to comment on those questions.

Is it good enough for us to ship an app that is mostly an empty skeleton by default that allows users to build & share their own boards? -I do think it would be helpful to start with a very basic board built in or have easy access to specific pages that you can choose to immediately incorporate. A lot of times, my device trials start with just figuring out how well they can navigate and comprehend, so being able to immediately try something would be a huge benefit.

I’m hoping that building boards could become a longer-term community effort, but it has to begin with building a functioning (albeit mostly empty) app. Does this approach feel effective? -Yes, but I could also see it becoming overwhelming as the app takes off. Maybe a rating/filter system for custom board would be helpful -like topic, age range (pediatric vs adult), real pictures vs images.

[BCBABritt]

Want to start off with saying as a BCBA I view my role in speech as a support rather than a driver. Which is why I don’t feel comfortable giving feedback on those aspects. What I would suggest is perhaps a quick ‘info’ tab for adults to read who aren’t familiar how to support an AAC user. Things such as, never take their AAC away from them, don’t turn down the volume, etc. and then perhaps one aimed more towards collaborative professionals like myself, written by SLP for us to review and ensure we are setting our clients up for the best success.

Disregard if this an awful idea, I’ve just seen too many technicians not get proper training and/or adjust button placement, visuals, etc. The only reason I am semi-aware of these things is because I had directly worked with an SLP for 3 years.

[PullJosh]

Sounds like we definitely need to ship a default board with some core vocabulary at minimum.

Does anybody know if there are existing core vocabulary boards that are legally free to use? Or is there anybody who would be willing to help write up a list of core vocabulary (while being very careful not to plagiarize from existing apps or make any lawyers angry)?

[myVoice]

Check out Project Core! Project core

[Gnat]

I have pulled out a few points, there have been a lot of good ideas flowing from others so I didn’t see the need to comment on everything.

• Question: Should tiles be able to do anything else?

Press and hold to change grammar. I have a friend who is an author. I have had her present her work to large groups which have included Q&A. I can see how it would have been useful for her to have some preprepared answers to predicted questions. It was also important for her to use more formal grammar in these situations. This is possibly part of the Gestalt vs Analytic language discussion.

• Certain tiles can be configured to speak immediately without being added to the message bar (Question: Is this a useful feature?)

This would be very helpful, especially for the general chat words/ phrases.

• It should be easy to have multiple different boards installed & switch between them

I have often wanted a simple “Yes, No, Other” page for those times where people just need the simple options without much clutter to help reduce the cognitive load.

• Question: Would a built-in drawing mode be useful? It would allow users to draw icons directly within the app, possibly with Apple Pencil support.

This could be very useful. An option (if not popular with the majority) might be to provide the dimensions and optimal DPI of a tile so that this can be done in apps like Procreate and transferred in.

• Question: Would using emojis as symbols be a helpful option?

As I use emojis in my text and email communication, having it as an option is a good idea.

• Question: Would AI icon generation be useful? (Type a word and it makes an icon for you)

I am cautious with AI myself. I have concerns around how much information can be accessed by AI, and the privacy related to that.

• Question: Are there any other image creation methods that would help?

Is screen shot already possible (would this require an editing step?)

Users should be able to customize audi

• By using a built-in AI voice

Question: How important is it to allow customizing the gender/accent/etc of the voice?

Is it possible to have a not AI voice option? I understand this might possibly increase the size of the app. Again, this is just from someone who is AI cautious.

As an Aussie I think it very important to be able to choose an accent and gender/ pitch characteristics of the voice that matches my environment.

• By recording themselves speaking

Question: Is this actually useful, or is it weird if the speaking voice is different for different icons because it was recorded by different people?

I can see how sometimes people have unique words and labels for things. If this is not a difficult addition, those who can see a use can use it.

• • By uploading audio files (One AAC user messaged me explaining that they like to have sound effects and sound bites from movies available for fun communication)

I love this idea. Many years ago, when iPad’s were new, I had changed my voice to be Yoda’s voice. Yes, it is novelty, but for us neurodivergent folk it is also something we enjoy. I know many students who would enjoy putting their favourite movie quotes on their device.

[christinahunger]

Hi Josh! These are all such great considerations. I can’t wait to see what you create! I’m an SLP and the biggest point I want to echo is the need for a ready-to-use vocabulary set instead of it being blank or mostly blank. I think it’s important to keep in mind that since this will be a free app, parents will likely often be the first ones to try it without a professional guiding them. If a parent is looking for a solution due to long therapy wait times, exploring options on their own, etc. they are going to be more likely to try a free solution rather than paying for a language software. So, it should require as few decisions as possible when opening the app and getting started.

Considering the app’s longevity, having established vocabulary set(s) also makes sense. If every user’s device has vastly different words and layouts, it becomes very challenging for the professionals who support AAC users. It’s normal to have some customization on each person’s device, but most of the words should be in the same place across users.

I love the idea above from BCBABritt of including a quick intro guide to AAC best practices or tips throughout for the parents / other professionals who are learning as well. I’m happy to help create those if you go that route!

I don’t know if this is possible, but could parents take a short quiz at the beginning and the app recommends which grid size to start with? I know it wouldn’t be the same as having an SLP assessing and making clinical decisions, but it could possibly increase the likelihood of parents starting at an appropriate level? I’m picturing questions about the child’s age, whether they’ve ever used a device before, if they can point their finger to select something on mom or dad’s phone, if there are any motor limitations, etc. I’d love to hear other SLP’s thoughts on this idea. And again, happy to help advise on what these questions would be and what the answers would signify.

[Melissa]

Hi! I’m an SLP with almost 15 years experience with the ID/DD population and AAC, currently working with medically complex high school students.

SMART WATCHES: Personally I haven’t come across a situation where that would be beneficial with my population (maybe adults with neurodegenerative disorders? But a phone app would support more robust language than a smartwatch). It’s important to have access to robust vocabulary with larger grid sizes, and pagesets on a device as small as a smartwatch would not be able to support that. It is not recommended to change grid sizes between different devices, which would interrupt the development of motor plans to access vocabulary.

What should happen if the same grid is used on a tablet and then on a phone? Just shrink it down but keep all the positioning the same?: Yes, positioning should remain the same regardless of screen size to support motor plans and access.

Certain tiles can be configured to speak immediately without being added to the message bar (Question: Is this a useful feature?): I think it can be useful! The people I work with have intellectual disability in addition to their developmental disabilities and do not use typical syntax on their speech-generating devices. We generally do not work on Subject-Verb-Object sentence order. Preprogrammed “Quickfire” phrases on many of the SGDs I work with now do not add the phrase to the message bar, which hasn’t been an issue for my students. It may be different if someone does not have an intellectual disability and would want to generate language with typical syntax structure.

Would a built-in drawing mode be useful? It would allow users to draw icons directly within the app, possibly with Apple Pencil support: I don’t think it would be useful and I haven’t personally seen this feature on any SGD. I also don’t think including emojis would be beneficial and I’m not sure about AI-generated images.

If possible, you should include the ability to modify symbols to make them CVI-adapted for people with visual impairments (e.g., black background, high-contrast, minimal colors).

MOTOR PLAN: Another thing to consider is if you want an option for a “motor plan” system where it will navigate back to the core page once you select a vocabulary word (instead of staying on the same page). This is a principle used in systems like LAMP Words for Life and TD Snap Motor Plan.

How important is it to allow customizing the gender/accent/etc of the voice?: Very important! I work with high school students and many have chosen to use the pitch and speech rate modification features to slightly change their chosen voices to match what is age-appropriate as they’re getting older. In my experience, users enjoy having personalized voices. It also helps differentiate voices in a place where there may be many AAC users. Some features that exist on SGDs now include child, teen, adult, geriatric, male, female, Spanish-accented English, and different emotions.

By recording themselves speaking Question: Is this actually useful, or is it weird if the speaking voice is different for different icons because it was recorded by different people?: I think the feature is important to include if there is something that is programmed that would not sound “right” using an SGD voice, like if it’s in a different language. I’ve done this for students who have a different home language (e.g., Arabic or Soninke that’s not readily available on SGDs) when we want to include social phrases and greetings, and we have a native speaker record those messages to use on the SGD.

Would it ever be useful to show text on the icon in just one language at a time?: I’ve never seen an app that shows the text of more than one language at once. I don’t know how useful it would be to list the text of both languages in the message window, but the verbal output is only in one single language? Generally you “toggle” between languages on devices when you used pre-built pages, and wouldn’t use two different languages at once. You can program words/phrases in another language within a pageset, but the voice output won’t sound native unless you record it.

Would it be helpful to let users set language-specific boards, or does switching languages just mean keeping the same board but in a different language?: It depends on the language. Ideally vocabulary, regardless of language, should be in the same location (which will happen for common languages like Spanish). But I’ve seen some apps that will reverse the toolbar locations and grids if the language is read from right to left (e.g., Arabic).

GESTALT LANGUAGE PROCESSORS should have access to both high-frequency, personally-meaningful gestalts (phrases) that can be programmed, as well as a core vocabulary “home” page with folders of fringe vocabulary. Phrase based language should not fully replace single words, but should be programmed in alongside a flexible and robust language system.

CORE WORDS should always be in a fixed location on the “home” or “core” page. Sometimes it can be helpful to have a row/column of core words that remain visible no matter where the user navigates. This reduces the need to backtrack to access high-frequency words like “go,” “stop,” “want,” “more,” “help,” “not,” etc. Otherwise there should be a icon somewhere on each page that allows the user to navigate back to the core page with one hit.

Instead of forcing a word to exist in only one place, frequently used fringe words can be accessible from multiple relevant categories (e.g., pizza might appear under “Food,” and on an “About Me” page if that’s something they love). This

TOPIC BASED PAGES: While core vocabulary remains consistent, temporary or context-specific pages (e.g., vacation, academic unit, holiday, doctor’s appointment, etc.) can be added and removed without disrupting the user’s established motor patterns.

ALTERNATIVE ACCESS METHODS: You should also consider users that rely on alternative access methods (e.g., switch scanning using a Tapio interface with an iPad, or eye gaze tracking). Newer iOS does offer eye tracking using the front-facing camera, but it may not be sensitive enough to use with an AAC app (I personally haven’t seen this work yet and typically the user will need separate eye gaze hardware).

SCANNING is an alternative access method used on an SGD that allows someone to select icons/cells without directly touching the screen. Instead, the device systematically highlights available options, and the user activates a switch (or another input method) when the desired option is highlighted. Scanning is commonly used by users who cannot reliably access a touchscreen due to motor impairments. Ideally an app should support multiple scanning methods (automatic, step/manual, inverse, row-column, group-item, etc.) to accommodate different motor abilities and preferences, allow users to adjust scanning speed, dwell time, auditory cues, and visual highlighting, and minimize the number of scanning steps required to reach commonly used vocabulary. TalkSuite is a good app to learn about to see how scanning can be integrated into a high-tech SGD.

[Melissa]

(I’m also an SLP!)

I think it’s a great idea to develop some sort of tool that guides parents to a more appropriate grid size. In my experience, people generally start with lower grid sizes because they don’t want to overwhelm the user, but it inadvertently limits language access, and will make it more difficult to change to a larger grid size when they’re “ready” because they will have to learn new motor plans. It’s much more beneficial to start with a larger grid and hide icons as needed.

I also added some info about alternative access methods in my reply, but I feel like trying to integrate eye gaze using the front facing camera (which is possible on newer iOS versions, but I haven’t seen it work yet with any apps I’ve tried) or switch-scanning may be a bit complex for the purpose of this project? I know TD Snap does have a scanning feature, but I’ve only ever used TalkSuite with my students that access their SGD that way. The language organization systems within the apps themselves are SO different.

Maybe the user gets prompted with a question about their access method when making a page set? If they use direct selection, it can be a more “traditional” organized core-page with topic pages, quickfires, etc. And if it’s programmed for a user who access an SGD indirectly via scanning, it would be a separate pageset similar to the folder system in TalkSuite? Just a thought!

[BGIOC]

Question: Would a built-in drawing mode be useful? Yes but also the option of uploading pictures such as photos.

Question: Would using emojis as symbols be a helpful option? Some would be (such as emotions and thumbs up or down) but not everyday items.

Question: Would AI icon generation be useful? (Type a word and it makes an icon for you) This would be helpful especially for parents and teachers that may not be as tech savvy.

Question: Is it helpful to support the open board format? Yes

Question: Would it be helpful to let users set language-specific boards, or does switching languages just mean keeping the same board but in a different language? I think language specific boards may be more helpful because usually they speak a language other than english at home and there will be lots of words and phrases used that will not be used at school.

Question: Are there other features we need for excellent multi-lingual support? Maybe being able to translate one word or phrase on the board if needed.

[MsKristy]

I 100% agree with BCBABritt that training for the user’s support team is crucial for success. As an SLP, I have often encountered parents and staff who are so worried about making mistakes that they don’t attempt to learn how the device works or they just don’t know how to engage with an AAC user. This means the user must learn the app with less modeling and less support or “buy in” for their communication method.

We have to make it easier and less stressful for the communication partners if we want the user to be successful.

Suggestions for Training Support Users

It would be helpful if the user could share their profile for training and increased modeling of the app. Comments could be added to the user profile as long as it fits within the Communication Bill of Rights and additional user centered guidelines.
Example: “Billy is loving the addition of classmate names and photos! Let’s give him the opportunity to call role this week?”

Location services with built in notifications to the supporting team’s device listing suggested opportunities for engagement.

Location: Fast Food Restaurant

Suggestions for engagement:

What type of dessert do you want?

I am really hungry today!

It could allow the support user to select ‘speak it for me’ or guide the support user to create the sentence.

Create a replay option of the user phrases so supporter users can also learn where the icons are located. Could be restricted for user privacy. Example:

User states “I like dolphins”. The Support User is guided to “I like (sea animal page).”

I would love to help with Support User training development!

[PullJosh]

I’ve been continually updating the original post based on all the amazing feedback here. I’ve learned about a ton of little details, but y’all have also changed my mind in one big way.

I was originally thinking that as developers (not SLPs), we should create a blank app and allow SLPs to fill it with useful vocabulary and share those boards with clients/the public.

After hearing your feedback, it’s obvious in retrospect: That’s way too much of a hassle!

To be genuinely useful, we need to ship an app that’s usable right away. I also really love @MsKristy’s concept of built-in help and support. It makes sense that getting started with an app like this is super challenging, and especially if we’re trying to fill a void for families who have been struggling to get started (perhaps because of cost or insurance headaches), we need to make it as easy as possible.

Going forward, we definitely need to work with SLPs (and maybe someone familiar with copyright law) to develop the default board.

Thank you so much for all the feedback, and please keep it coming!

[StephanieS]

@PullJosh

In-app training & help

• Feedback received: In practice, families often struggle with supporting AAC-using children because parents/guardians aren’t familiar with these tools themselves and are learning alongside their child. Built-in help and support would dramatically increase the practical usefulness of the app.

IIn-app training and pop-up support sounds great, especially for caregivers just getting acquainted with the app or device. Starting out with a new SGD can be intimidating, so built-in training and support would go a long way toward driving adoption and buy-in. One consideration is storage — I’m not familiar with the development side, so this may be negligible, but it’s worth flagging. I’d also recommend diversifying the training and support content based on who’s interacting with the device. The training and support needs of a parent of a nonspeaking or minimally speaking child often look very different from those of a caregiver supporting someone with recently acquired aphasia.

@PullJosh

• Users should be able to turn on multiple languages. When they do so, a language picker should be added to the UI that allows them to quickly switch the speaking language

When programming a device in TouchChat for someone from a bi/multilingual household, I typically build out the person’s vocabulary in their primary language and then duplicate it in their secondary language. TouchChat has an enablement feature that lets users toggle between vocabulary sets with the press of a button — a similar feature here would be a great way to support multilingual users. One improvement I’d suggest over TouchChat’s approach is a vocabulary set translation option, since tailoring vocabulary across secondary languages is often very time-consuming.

@PullJosh

• By recording using the built-in microphone
• By uploading audio files (One AAC user messaged me explaining that they like to have sound effects and sound bites from movies available for fun communication)

This is great, especially for younger users and gestalt language processors (GLPs) — totally necessary, in my opinion. With AAC/SGDs, the goal is for the device to reflect and expand on the language the individual already has, whether that’s purely receptive language, receptive with minimal expressive language, or receptive language paired with gestalts.

@PullJosh

Users should be able to share their customized boards

• It should be possible to download the board as a file to be shared via email, Google Drive, Teachers Pay Teachers, AirDrop, etc.
  • Question: Is it helpful to support the open board format?
• It should be easy to generate URLs or QR codes so that others can add your board to their device instantly

Agreed, 100%. Having a vocabulary set that can be easily shared and saved outside the primary device is critical. If a device is damaged or a vocabulary set is deleted without being backed up elsewhere, therapists and caregivers are left having to reprogram the entire device from scratch, often compromising treatment time.

@PullJosh

• It should be easy to have multiple different boards installed & switch between them
• Feedback received: Distinguish between core word boards and vocabulary boards.
• Feedback received: Have a setting that can be turned on so that every time a vocabulary word is selected the user is automatically brought back to the core word screen (for easier motor planning)

Transitioning from single-word or sentence-level device-based communication into conversational communication can be genuinely difficult. Even with a core word page in place, I often find myself building out separate conversational pages for specific contexts — the doctor’s office, the playground, the grocery store, and so on. TD Snap has a similar feature for this, and I believe Lingraphica is developing something comparable. Most major device and app producers offer device trials, so it would definitely be worth exploring what’s already out there to see how others are approaching this.