New Study Suite Feature: AI Text-to-Speech 1024 682 Visual Veggies

New Study Suite Feature: AI Text-to-Speech

AI is now part of Study Suite

I’ve got some really exciting news, especially for our auditory learners and audio-visual learners.  The Study Suite software now includes a new feature:  Text-to-Speech (TTS) via AI

There was a feature in the software where the Library articles and the Practice Exam explanations can be read aloud.  However, this feature was utilizing the computer’s built-in voice synthesizer.  If you’ve ever used it, you would have to agree it is very robotic-sounding.

I’ve had on my software features to-do list (I call it my wish list) a task for myself to eventually work on improving the Speak function in the software.  I kept putting it off as I didn’t think many students actually used it, and those who did use it never complained about how unpleasant it was.  Others may not have thought it was that bad, but I definitely thought it was unpleasant, and that’s stating it nicely.  Earlier this month, I just started Googling “text-to-speech APIs”, just to see what’s out there and how complicated something like this would be.  ICYDK, an API is a service in which an app reaches out to with some data.  Then, other data is sent back to the requesting app.  In the case of this new feature, text is the data sent from the software to the API service.  The service converts the text to audio, and then sends this raw data back to the software, where it is then recompiled and turned into audio.

API stands for Application Programming Interface.  And in case you didn’t know, ICYDK stands for “in case you didn’t know” :)

Ok, enough acronym talk.  We’re getting off track.  Let’s get back to the AI story.

Through my research, I found OpenAI (you know this better as ChatGPT) has a TTS feature one can utilize via API, so I started playing around.  Turns out, it wasn’t that difficult, and the result is amazing IMO (IMO = in my opinion… sorry another acronym!)

Take a listen for yourself!

AI can be used for so much

When we think of AI (artificial intelligence) and ChatGPT, you might automatically think AI is only good for textual capabilities.  For example, you give the AI some prompts, and it searches its ever-growing, ever-learning database and provides an answer.  You might have asked it to write a short story, craft an email reply, or answer a question.  The response from AI is not always perfect.  The response is reliant upon the available information on the internet and its training regimen.

But we’re not limited to only text responses.  Digital artists might use AI to create any image imaginable.  Software developers can use AI to help with problematic coding issues.  I actually used AI a couple of times working on this TTS feature!  And as we see, AI can even be used to turn regular text into beautifully-sounding audio.

Details about the new software feature

  • Text-to-speech via AI is now available in the Study Suite software!  (desktop only at this time)
  • You can find it in the Library articles, the Practice Exam explanations, and the Exam Review section.  Click the Speak button on these screens.
  • The audio output is that of artificial intelligence (AI) and not the voice of a real human being.  (OpenAI asks that this statement is made when using their service)
  • You can choose between a male voice and a female voice in the Settings.
  • The process can take several seconds for the audio to start playing.  This is because the text needs to be sent to the internet, converted to audio via AI, and then transmitted back to the app.  To help with the speed, the recordings will be saved, so the next time you want to listen to the article/explanation, it will be instantaneous.
  • You can now pause, restart, replay, and scroll through the audio.  Students were unable to do this with the old built-in voice synthesizer.