App Review: Predictable – Part 1

by Cyndee Williams Bowen

The folks at Therapy Box have recently updated their Predictable 4 Augmentative and Alternative Communication (AAC) app. I have explored Predictable 4 for a couple of months now, and I find it most impressive for several reasons.

Intro to the App

Predictable 4 is intended for users who have lost the ability to speak but demonstrate relatively well-preserved cognition. Survivors of stroke, traumatic brain injury, or people living with progressive neurological disorders such as Amyotrophic Lateral Sclerosis (ALS) or Multiple Sclerosis (MS) may benefit from the straight-forward instructions and easy to use, intuitive features offered by this app.

The look and functionality

The display screen in Predictable 4 is uncluttTB-Predictable4-IntroPgered and visually pleasing. Color schemes can be customized according to user preference. The purple message screen with a 30 point, white text font is the combination that worked best for me.




Text can be enterPredictable4 - IntuitiveWriting1ed using the on-screen keyboard, a separate keyboard, or via the app’s writing feature, shown here.  Notice that the program is learning my first name and offers it as the first intuitive choice for the sentence I am creating.




Predictable4 - IntuitiveWriting2But my last name was not offered as a choice, so I had to write it out. This is a nice feature for people who prefer to write vs type. I also envision it as a functional therapy activity for people with expressive communication deficits that impact both speech and writing.




TB-Predictable4-Cursive MicroPredictable 4 also recognizes cursive entries. My stylus limited my ability to produce a small, cursive message, but all sizes of print I tried were recognized. Micrographia is a problem often experienced by people with Parkinson’s disease, so I am especially happy to see that the program handles small print well.


Options for news, messaging, and participation in social media

One advantage of choosing an iPad, tablet, or computer as an AAC tool is the ability to utilize it for multiple functions. The device serves as a speech generating device (SGD), but it may also be used to conduct business, send messages, read books, watch movies, listen to music, and connect with others via social media. The Use icon on Predictable 4’s message screen/keyboard offers choices that facilitate access to information, editorial tools, and engagement in social media. The following 12 choices are offered: Add Phrase, Copy, Paste, Email, Facebook, Twitter, Skype, News Feeds, Message, History (previously produced utterances), Favorites, and Print.

Thus far I’ve focused on the visual and functional aspects of Predictable 4. These features are impressive, but I’ve saved the best for last.

Voices and ModelTalker*!

The sound of an individual’s voice is a defining feature – like eye color, shape of the nose, or complexion. People are often as attached to the sound of their own voice as they are to other features. They may grieve its loss as much or even more than the loss of a limb.

Current AAC apps and SGDs are usually equipped with voices that can be TB-Predictable4-NuanceVoicespersonalized by gender, age group, dialect, and sometimes even  pitch range.  Although the quality of these voices is improving, they typically sound a bit robotic, generic, and unlike the individual’s natural voice. Predictable 4 offers a variety of high quality voice selections in multiple dialects, but the most exciting feature of the program is inclusion of the beta version of a new voice creation engine called ModelTalker!

Predictable 4 and its sister program, Chatable 2 are the first AAC apps to include a beta version of ModelTalkerModelTalker creates a personalized voice from samples provided by the user. You can visit the ModelTalker website to explore on your own and even listen to a demo of text you provide. The voices produced still sound robotic, and quality of recordings provided by the user will impact the final result, but the test voices created from normal voices under ideal acoustic conditions manage to capture the essence of the speaker’s own voice. The goal of the program is to build voices that sound as close to the user’s natural voice as is technologically possible. It’s not perfect, but they seem to be on target.


Part 1 of this blog has introduced you to Predictable 4, an AAC app offered by Therapy Box, and its inclusion of ModelTalker, a new, exciting development in speech synthesis. Creating my own individualized voice using ModelTalker will take some time. I will have to register with the company; purchase a higher quality microphone (because I’m a bit of a perfectionist); and then create and submit the best quality speech samples I can manage in my office environment. I am off now to contact the company and begin the process! I will pick up there and let you hear the results in my next blog in this series: App Review: Predictable – Part 2.



CyndeeBlue2Cyndee Williams Bowen owns Bowen Speech-Language Therapy, LLC in Clearwater, Florida. She provides collaborative, creative, quality services to adults and adolescents with speech, language, voice, swallowing, and related disorders. Follow the links to contact her and visit the Bowen Speech Website for learn more about Cyndee and the services she provides.

One thought on “App Review: Predictable – Part 1

  1. Pingback: App Review: Predictable 4 – Part 2 | Bowen Speech Blog

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s