Two Spaces and a Hard Enter: Pro Tip for Text-to-Voice AI Tools in ID

When using the wonderful new text-to-speech tools in Articulate Storyline 360, specifically (this applies to all audio tools so I happily generalize), you should be using some cushioning techniques for your text-to-speech audio because you don’t always know how fast the Internet connection is for your learners…some developers automatically add a 3-5 second cushion at the start and end of their audio clips. I subscribe to this line of thinking wholeheartedly, but in practice, I end up customizing more than what gets automated, and I have no problem with that because the end product is just sooo much nicer).

Articulate Storyline, like most instructional design and eLearning tools, will allow you to take slide notes, or the text to be narrated, and narrate the text with one of several fantastic new AI voices. The tools have been there for a while, but they’ve been noticeably “different” until recently; not in a good way. Now? WAY different, and for the better. The voices are awesome, and you just have to learn how to work with them (how do I get you to say Vay-K without sounding so snobby, Robo-Chaz?) as they, in turn, learn you…it’s neat and a little spooky, perfect to lead you into the Halloween season. You know that's right.

So, ahem, please forgive the tangent: Two Spaces and a Hard Enter (which is basically two line breaks) by entering two spaces and hitting the Enter key after entering the text to be translated has been typed or pasted. This will add just enough buffer to prevent the narrator’s voice from getting cut off at the end of the audio clip. I notice it happens quite a bit when using bulleting, which should make sense because you are adding additional formatting (think noise) and code behind the scenes, introducing more room for error, while creating those lovely opportunities to delight and educate in differently impactful ways.

All is manageable, but you must keep your code clean, label and note anything you decide to get fancy with behind the scenes. It’s the same thing as not being an jerk and commenting your code for programmers (I used to teach Intro to programming courses and was sometimes harsh on students for not commenting their great work, and for not self-attributing…don’t be invisible; let people know you did that!). Instructional Designers don’t have the same professional respect for code that programmers do, so you really need to help your peers and nudge them to document their stuff, and show ‘em how. They want to learn, and you can remove that barrier for entry.

I cannot tell you how often I must gut and start over with work from others that has no explanation of how to engage with it, or logical/any entry points to alter it. Usually, the work is kind of bad anyway, so you won’t care and are happy to get a blank slate, but I get the occasional surprise where someone did something ridiculously brilliant, I cannot replicate (are you telling me that I all I gotta do is just make this pretty? No, it can’t be that good, can it? The golden light from the briefcase of riches! Sometimes, it really is, and it is soo cool when it is better than your work and you can learn from it), with no access to the person or that level of expertise to do anything with it, without a lot of time invested to make it happen. You don’t mess with billable hours, so buh-bye...mmm, still, buh-bye.

You cannot be oblivious to the fact that your work leaves a footprint; you may get the opportunity to engage with people who’ve worked with your stuff, and it all feels pretty great when you are putting your all into your work, because that’s how you can relationship-build and nurture a community. I work with a bunch of people who like to get my files; that feels plus-ultra awesome and is the type of validation I find priceless and you should too. It all helps you to know that you are baby-stepping in the right direction.

The example I was thinking of still annoys me because I rewrote the whole thing and liked the original’s functionality a lot better than what I did to mirror it (read: I think the other person used 10 lines of code to make something work that took me 100); rare, so it stuck with me, but I learned a lot from that experience, and maybe you can learn something from this too. Anyway, that’s your tip to add a little cushioning you can tweak to optimize your workflow and the user experience while working with AI text-to-speech tools, that are finally living up to some of that hype.