Welcome back, fellow SLPs! In our last two posts, we tackled the crucial topic of AI and Clinical Data, as well as AI Truths & Myths. We learned that responsible AI use starts with safeguarding our clients’ privacy and understanding how AI works in practice.

We’ve debunked some common myths and explored the complex questions surrounding AI and data security. Now, let’s pull back the curtain and take a look at the mechanics room. In this post, we’ll demystify how AI tools actually work, providing a clear and simple explanation for SLPs.

Artificial intelligence can sometimes feel like magic, especially when it comes to generating coherent text for notes, therapy ideas, or reports. You type in a prompt, and seemingly out of nowhere, clinically relevant content appears. But behind this impressive capability lies a fascinating, albeit complex, system built on specific AI principles.

As SLPs, understanding the basics of how AI tools function is essential. It can not only alleviate some of the mystery, but also empower us to use these tools more effectively and with realistic expectations. So, let’s peek under the hood and demystify the mechanics of AI.

The Foundation: Large Language Models (LLMs)

To truly understand how AI works, you must first understand its base. At the heart of most AI tools you encounter that generate text are Large Language Models (LLMs). Think of these LLMs as incredibly advanced prediction machines for language.

They’re not thinking, feeling beings; instead, they are sophisticated algorithms. They are often built on a “transformer” architecture, that have learned to predict what word or sequence of words is most likely to come next. This ability to predict is the foundation of how they generate paragraphs, articles, notes, and even initial drafts of reports.

The Training Process: Learning from Text Patterns

The secret to this predictive power lies in the training process. LLMs are trained on truly massive amounts of text data – billions upon billions of words scraped from books, articles, websites, code, and just about any other digital text you can imagine. For models used in healthcare or clinical research, this could also include vast quantities of de-identified research papers, medical journals, and clinical guidelines.

During this training, the AI doesn’t process this raw text directly. Instead, it goes through a crucial step called tokenization.

What is Tokenization?

Tokenization is the process of breaking down the text into smaller, manageable units called tokens. These tokens can be whole words (“the,” “client”), parts of words (“-ing,” “re-“), or even individual characters or punctuation marks. The way text is tokenized can vary between different AI models, but the fundamental idea is to convert human language into a numerical format that the AI can process.

Think of tokenization like the raw data stream in the movie The Matrix. Instead of seeing words and sentences, the AI sees a constant flow of these individual numerical tokens. This numerical representation is the fundamental language that the AI understands and manipulates.

Each unique token in the AI’s vast vocabulary is then assigned a numerical ID. Therefore, AI operates at this granular level of tokens and learns statistical relationships between them. It doesn’t typically “hold” or store entire books, articles, or clinical records as copies. Instead, it retains the learned patterns and probabilities of token sequences.

How is Tokenization Used?

Once the text is tokenized and converted into these numerical representations, the AI analyzes this vast ocean of language, identifying patterns. It learns the statistical probability of certain tokens appearing together. This includes the common structures of sentences and paragraphs, and typical ways humans express different ideas in various contexts (e.g., academic writing, informal speech, or clinical documentation).

For example, the AI learns that after the tokens “Client” followed by “demonstrated,” the token “progress” has a very high probability of appearing next in a clinical note. It learns the statistical relationships between diagnostic terms and common intervention strategies.

It’s all about recognizing these statistical relationships between these numerical tokens. It’s crucial to remember that the AI doesn’t “understand” these relationships in the way a human clinician does. It doesn’t grasp the clinical meaning, the client’s emotional state, or the unique context of a therapy session. It simply recognizes that, based on the numerical sequences it has processed, certain token sequences are more probable than others.

Mechanics of AI Text Generation: Prediction in Action

So, how does this learned knowledge translate into generated text that can assist an SLP? Let’s discuss the AI principles at work. First, you give an AI a prompt – a request, a question, or even just a few keywords (e.g., “Draft a SOAP note for a 5-year-old with articulation goals for /r/”). Then, the AI first tokenizes your input.

After that, it processes its learned patterns and the numerical IDs of these tokens. It then predicts the most likely next token (represented by its numerical ID). This prediction is heavily influenced by the parameters of the LLM. Think of these parameters, often billions or even trillions, as the internal “knobs and dials”, like the Oz Wizard. The AI adjusts these during its training to learn the strength of the statistical relationships between tokens. The more parameters an AI has, generally the more nuanced and sophisticated its predictions can be.

The Output of the AI Machine

The AI takes that newly predicted token. It compares the original prompt and the token it just produced. Then it predicts the next most likely token, again relying on its vast network of parameters. This process continues, token by token, until the AI determines it has reached a stopping point. Finally, these numerical tokens are converted back into human-readable text.

Think of it like a super-powered version of the autocomplete feature you might use when typing. However, it operates at the level of fundamental text units, and is guided by an incredibly complex web of learned relationships stored within its parameters. The “creativity” or “clinical relevance” we sometimes see in AI-generated text isn’t true clinical reasoning in the human sense. Rather, it is the result of the sheer vastness of the data it has been trained on. It has learned the intricate statistical relationships between these tokens. It is all governed by its numerous parameters.

What AI Isn’t: Understanding the Clinical Limitations

Finally, it’s vital to reiterate what AI isn’t, especially from an SLP perspective. Even understanding the mechanics of AI, its human-like responses can often muddy the waters.

• It isn’t conscious. It doesn’t have emotions, personal experiences, or clinical judgment to draw upon.
• It cannot understand or provide the nuances of human communication. Sarcasm, irony, or the subtle emotional context of a client’s utterance are often difficult for AI to understand. They are even harder for AI to create (don’t ask it to generate jokes).
• It is limited in understanding highly nuanced linguistic elements like prosody, intonation, or specific phonetic realizations (e.g., truly hearing and analyzing a distorted /s/ or a lateral lisp). These areas are critical for SLP assessment and intervention, but challenging for current AI to grasp fully and accurately.
• It cannot build genuine therapeutic rapport or provide the empathetic connection essential for effective therapy.
• It cannot make a differential diagnosis or independently formulate a comprehensive, individualized treatment plan based on dynamic clinical interactions.
• AI can’t handle very large chunks of information. This is a significant limitation to remember, especially when working with long documents. More advanced and often paid AI versions, which are hopefully HIPAA-compliant, can process larger numbers of tokens.

Its output, however impressive, is ultimately based on the patterns it has learned from the text it has been trained on. You must remember that it is processed at the level of tokens and statistical probabilities, not genuine clinical understanding. You MUST always double-check output for accuracy, completeness, and bias!

Looking Ahead: Understanding AI as Your SLP’s Toolbox

Understanding the core mechanics of AI—that it’s a sophisticated text prediction machine built on statistical patterns—is the key to unlocking its potential without falling for the hype. This knowledge empowers us to see AI not as a magical entity, but as a powerful tool in our clinical process. It’s an assistant that can handle routine tasks, potentially allowing us to save precious time and redirect our energy to where it matters most: our clients.

Now that we’ve peeked under the hood, what are your initial reactions? Has demystifying AI’s inner workings changed your perception of its potential uses (or limitations) in SLP? Share your thoughts in the comments below, and then weigh in with your perspective in our quick poll! Results will be shared at the end of the series.

AI & SLPs Poll 3

I’d love to hear your thoughts after reading this post! Please answer these 4 questions; your valuable input will help shape my series’ final insights. To keep the poll fair and ensure unique responses, a Google account sign-in is required. Be assured, your email address is neither collected nor visible to me.

The AI & SLPs Series: Your Comprehensive Guide

Welcome to the AI & SLPs Series! Over the next eight weeks, we’ll delve deep into how Artificial Intelligence is shaping the world of speech-language pathology. Here’s what you can expect:

• Part 1: AI & Clinical Data Privacy
  • This foundational post explores AI training data, client privacy, and HIPAA compliance for SLPs, including the non-negotiable role of BAAs.
• Part 2: Separating Truth vs Myth 
  • We debunk common AI myths in SLP practice. Get a realistic understanding of AI’s true role and capabilities.
• Part 3: How AI Tools Work (current post)
  • Get a clear, jargon-free explanation of how large language models function. Understand their capabilities and limitations.
• Part 4: AI for Clinical Spark & Efficiency
  • Discover ethical ways to use AI. Brainstorm, overcome planning hurdles, and refine non-clinical communications.
• Part 5: Mastering Prompts 
  • Learn prompt engineering. Communicate effectively with AI models to get tailored, useful results for SLP needs.
• Part 6: Compliant Platforms & Tools
  • This post guides you through AI tools. Learn key factors for ethically and compliantly selecting platforms for your SLP practice.
• Part 7: Ethical & Responsible Use
  • This crucial post delves into broader ethical responsibilities for SLPs using AI. It covers principles beyond data privacy.
• Part 8: The Future of AI
  • This concluding post explores emerging AI trends and future possibilities in Speech-Language Pathology. Prepare to adapt, innovate, and lead responsible AI integration.

Stick around as we keep figuring out this whole thing together, giving SLPs the knowledge they need and helping us all find a balanced way to think about AI in the future of speech-language pathology. There is a lot of gray area and strong opinions, and I hope I can provide some facts to help you make informed choices that correspond with your own values.

Keep on clickin’!

Social Media Icons: designed by rawpixel.com – Freepik.com

The post AI & SLPs Series, Part 3: How These Tools Actually Work appeared first on Mrs. Speech Online.

Then life threw a curveball, and I resigned from the schools.  I ended up taking a position with a teletherapy company and providing services to various schools.  I was hesitant at first, but after working as a teletherapist for six months, I can say I completely LOVE it!

Comfort

Need I say it? Yoga pants!!!  Business on top, (slumber) party on bottom, am I right?!!  I had a 40 minute, each way, commute before, now I walk down the hall to my office.  I can run to the bathroom or to get a coffee refill or a snack whenever I need.  And have I mentioned no boogers on my stuff?!  I haven’t had a sick day from any child-borne illnesses so far!

Variety

Not all teletherapists have the variety I do, but I have students ranging from mild middle school students to non-verbal kindergarteners.  I enjoy the different caseloads, although it does mean a bit more prep as there is more diversity in goals and materials.  In the morning I get to talk about animal sounds, and in the afternoon I get to teach /r/, and the next day I’m teaching a great short story to work on language skills.

Use of Time

When I worked in the schools, a good part of my time was spent doing duty, attending teacher meetings, filling out paperwork required by the district, completing screenings I was unable to take any action on, performing before or after school requirements such as parent-teacher meetings whether or not I actually had any parents wishing to meet.  Since I was on a contract, I was paid by year, and these times were all under a blanket “duties of the position” statement.

As a teletherapist, I am paid for indirect time as well as direct time.  This differs somewhat from one teletherapy company from another, but from my, admittedly limited, experience, this is typical.  I only attend meetings that directly affect my students.  I feel like I am able to return back to my actual occupation!

Freedom of Scheduling

Along the same lines, I get to draw the lines of my schedule.  I don’t want to work past three?  Then I say I’m not willing to.  I only want to work 30 hours a week?  Then that’s what we shoot for.  I’ve also found that the schools that I work with are more willing to work with me as far as scheduling than I experienced when I worked in a district.  Perhaps I’m just in a sweet spot?  Time will tell.  But I’m loving it right now!

Less Drama

All those powerplays that you couldn’t keep out of in the schools are far less apparent as a teletherapist.  Because we aren’t physically there, we don’t hear the conversations in the halls or the teacher’s lounge. It’s much the same as being at a school part time, where you see your handful of kids and leave.  This, for me, was a fantastic load off my shoulders, because that kind of conflict stresses me out – and there was always someone who was running to me complaining about something.  Now, this can be a drawback as well, especially if you thrive on those connections.  But you can also make connections with your team of teletherapists.

Efficacy of Therapy

Now, do I think I can do therapy as well through a computer as I can in-person?  Yes!  Now, that being said, there are, of course, some caveats.  The method of therapy should be considered for each client just as the time, length and frequency of therapy should be considered.  Often, as my younger students have, a facilitator may be present to aid the client in attention and accessing the technology. 

Other clients may need a different service model that is more of a parent training model.  A setup that includes a separate camera that shows the student’s AAC device so you can see their key presses, or other modifications might be necessary.  With flexibility and creativity, there really is a lot more you can do in teletherapy than you ever thought!

Drawbacks

Ok, yes, I miss the goodies in the teachers’ lounge.  And I REALLY miss the hugs and actually seeing my kiddos.  Sometimes, it is challenging when I have a student that just can’t get a specific speech sound…but I’m not sure being in person would help that either.  Tech issues can be frustrating, when the microphone isn’t catching a nuance of sound or the connection isn’t good.  But there were also times when the hallway outside my room, or the room beside my room, or the restroom beside my room, were so loud that I had trouble then too.  So I might be comparing apples and oranges there.

I’m not necessarily writing this to convert any of ya’ll from in-person therapists to teletherapists, but definitely, should it come to you, don’t be afraid of it.  It could be your next great thing!

If you want to know more about teletherapy, or the company I contract with (starting on my third year with them), feel free to reach out!

If you are interested in the topic of teletherapy, you might also be interested in my comprehensive guide, “So You Want to Be a Teletherapist.” It covers all these topics and more, including setting up your workspace, navigating legal documents like BAAs, and building a secure and efficient digital practice from the ground up!

Until next time, keep on clicking, my e-speechies!

Social Media Icons: designed by rawpixel.com – Freepik.com

The post Why I Love Being a Teletherapist: From Skeptic to Superfan appeared first on Mrs. Speech Online.

Supercharge Therapy Planning:

• Personalized Materials: Struggling to find the perfect stimuli for a particular client? AI can generate engaging stories, activities, and exercises tailored to their specific needs and target sounds.
• Brainstorming Buddy: Facing a complex case? Use AI to brainstorm creative therapy approaches and explore different treatment options. Here, your expertise is crucial in selecting the most appropriate options based on your client’s individual needs and learning style.

Boost Client Engagement:

• Interactive Activities: AI can create interactive games and exercises that make therapy sessions more fun and motivating for clients, especially younger children. However, use your expert insight to ensure the difficulty level and format of the activity are appropriate for your client’s abilities.
• Real-Time Feedback: Imagine having an AI assistant that provides real-time feedback on a client’s pronunciation, intonation, or fluency! This can offer valuable insights and promote self-monitoring skills. Remember to analyze the AI’s feedback alongside your own observations to provide the most comprehensive guidance to your client.

Revolutionize Documentation:

• Automated Note Taking: Free yourself from tedious paperwork! AI tools can transcribe therapy sessions and generate draft session notes based on the audio recording. You can then review and edit them for accuracy, incorporating your clinical observations and insights.
• Speech and Language Sampling: Automatic Speech Recognition (ASR) tools can transcribe the recordings with high accuracy, saving SLPs valuable time. Additionally, AI can analyze the transcribed speech samples for various aspects. It can identify features like articulation errors, fluency patterns, and vocabulary usage. This should complement your human expertise, which remains crucial in interpreting the data.

Ethical Considerations:

• Focus on Collaboration: Remember, AI is a tool to enhance your practice, not replace your specialized skills. Use AI-generated materials as a starting point, then personalize them to fit your client’s specific needs.
• Data Privacy: Ensure any AI tools you use comply with data privacy regulations and handle client information securely. I’d recommend leaving out names or any identifying information.

Prompting Your AI Assistant Strategically:

To get the most out of AI tools, it’s important to provide clear and concise prompts that incorporate your expertise:

• Client Profile: Briefly describe your client’s age, communication goals, specific challenges they are facing, and their learning style. This information helps the AI tailor its output to your client’s needs.
• Target Area: Specify the area you’d like AI to assist with, such as creating articulation exercises (mentioning the specific sounds you’re targeting), developing social communication activities (specifying the desired social skill), or crafting reading comprehension worksheets (indicating the reading level and comprehension skill you want to focus on).
• Desired Style: Indicate any preferences you have for the AI’s output. Do you want the activity to be interactive, story-based, or game-like? Consider your client’s interests and learning style when making this choice.

Remember: You are the expert. 

Use your professional knowledge to evaluate and adapt any AI-generated materials to ensure they are appropriate for your client’s specific needs and goals.  The AI is a tool to empower you, not replace you.

Ready to Get Started?

Explore the world of AI resources available for SLPs! Many AI writing and speech recognition tools are readily available. Start by identifying specific areas in your practice where AI could be most beneficial.

By embracing AI as a collaborative partner, While keeping your professional knowledge at the forefront, you can streamline your workflow, personalize your therapy approach, and ultimately empower your clients to achieve their communication goals.

Envision this:

Sarah, a young client, struggles with sharing toys during playtime. You, the SLP, identify the need to create a social story to help Sarah understand the concept of sharing and taking turns.

Here’s where AI becomes your partner:

Prompting the AI: 

The SLP prompts the AI with the following information:

• Target Social Skill: Sharing toys
• Character: A young girl named Sarah
• Desired Tone: Positive and encouraging

AI Generates Story:

Sarah loves playing with her toys! She has a cuddly teddy bear, a bright red car, and a set of colorful blocks. One sunny afternoon, Sarah is playing in the park when she sees another child, Ben, sitting alone. Ben looks a little shy.

Sarah walks over to Ben with a big smile. "Hi Ben! What are you playing with?" she asks.

Ben mumbles something about having a toy truck but not wanting to play alone.

Sarah has an idea! "Would you like to play with my blocks?" she asks.

Ben's eyes widen. "Really? Can I?"

"Sure!" says Sarah. "We can build a big tower together! Do you want to put the red block on top?"

Ben smiles and carefully places the red block on top of the growing tower. They take turns adding blocks, giggling as the tower gets taller and taller.

Suddenly, another child, Emily, walks over and looks at their tower. "Wow! That's amazing!" she exclaims.

"Would you like to help us build even higher?" Sarah asks Emily.

Emily joins in the fun, carefully adding more blocks to the tower. They all work together, taking turns and laughing as they build a magnificent tower of blocks.

SLP’s Integration:

After reviewing the story to insure it is appropriate to the child, the SLP reads the story with Sarah, pointing out the pictures (ideally featuring a character resembling Sarah and toys she recognizes) and discussing the emotions of the characters. They talk about how Sarah shared her toys with Ben and Emily, and how everyone had more fun playing together.

Personalizing the Story: 

The SLP notices Sarah particularly loves animals and edits the story to mention her favorite stuffed animal, Mr. Snuggles, instead of a teddy bear. They can further personalize the story by adding details about the park (mentioning any specific equipment Sarah enjoys) or the type of tower they build (e.g., an animal tower if Sarah is interested in animals). Please be aware: personalization like this could constitute Personal Health Information under HIPAA. Do not share PHI in ANY AI platform without checking that it is HIPAA compliant! See more in AI & SLPs Series, Part 1: Clinical Data Privacy.

The SLP can also create follow-up activities based on the story, such as role-playing sharing scenarios or drawing pictures about taking turns using their favorite toys.

Benefits of AI in Social Story Creation:

• Overcomes Writer’s Block: AI can provide a starting point or inspiration for the SLP.
• Saves Time: AI can generate a basic story outline, allowing the SLP to focus on customization.
• Engaging Language: AI can create stories with a child-friendly tone and vocabulary.

This is just one example of how AI can be used as a tool to enhance the work of SLPs. Remember, the SLP’s professional knowledge and experience remain crucial in tailoring the story to the specific needs of each client.

Keep on clickin’!

Social Media Icons: designed by rawpixel.com – Freepik.com

The post Unleash the Power of AI in Your Speech Therapy Practice appeared first on Mrs. Speech Online.