Closed Captioning in Education — the DCMP

I need to start by saying, without hesitation: I believe Storyline is simply the best in eLearning development environments, and Articulate's dedication to accessibility is just one of the reasons. So please don't forget that as I go on to point out an area where it still needs a great deal of improvement: the closed captioning system. Since the implementation of AI voices I've been quite surprised to discover it fails to remove SSML markup, starts sentences inside captions and splits them at unpredictable, often awkward places within sentences. Captions are too small by default, and the opaque background quite simply is far short of best practice. Please hear me out. 

The Described Media and Captioning Program (DCMP) has put over two decades of research into their Guidelines and Best Practices for Captioning Educational Video, available at https://dcmp.org/learn/captioningkey 

To achieve captions that approach DCMP standards for "Quality Captioning" I still use the free open source program SubtitleEdit, available on GitHub https://github.com/SubtitleEdit. The software can be set to reflect line length and readability standards such as estimated rate (wpm), and to start new captions after end-of-sentence punctuation — even to automatically split lines at conjunctions, which is considered a must for achieving AAA success for those with reading challenges and cognitive disabilities. It even lets me save a slightly different configuration of these elements for French language eLearning I also produce. The arrival of an AI assistant seems to offer Articulate an opportunity to merge this knowledge into your products. 

I set the magnification in the Modern Player to 180%, (I also use the highly readable font, PT Sans Caption, but the default sans-serif is fine)  and with a plain text editor running as Administrator, Find the following line in "C:\Program Files\Articulate\360\Storyline x64\player\unified\html5\lib\stylesheets\mobile.min.css" and [same path to]\desktop.min.css: 

.caption p{margin:0}.caption>div{pointer-events:none;width:100%;text-align:center}.caption>div>p{display:inline-block;pointer-events:none;padding:10px;text-shadow:-1px 0px #000,0px 1px #000,1px 0px #000,0px -1px #000;color:#fff;border-radius:4px;background:#31373a;margin:0 0 10px 0;padding:10px;text-align:center;line-height:1.5}

and then Replace it with:
.caption p{margin:0}.caption>div{pointer-events:none;width:100%;text-align:center}.caption>div>p{display:inline-block;pointer-events:none;padding:10px;text-shadow:-1px 0px #000,0px 1px #000,1px 0px #000,0px -1px #000;color:#fff;border-radius:4px;background:rgba(33, 33, 33, 0.81) !important;margin:0 0 10px 0;padding:10px;text-align:left;line-height:1.15}

This assures the captions are left-aligned per best practice, readable by those with vision challenges (now set within the Player, not the above CSS), and translucent, as the DCMP recommends in order that content isn't obscured. 

Where I live, my own work in eLearning is public-facing and therefor overseen by the Accessibility for Ontarians with Disabilities Act (AODA) and the Ontario Human Rights Code. As I work for one of the agencies mandated and defined within the latter, I am heavily invested in this knowledge. I'd like to offer these observations as enhancement suggestions (I know there's a separate area for that, but don't hesitate to remind me precisely …where!?) I'm happy to provide fellow Heroes with greater details and answer any questions. 

Thanks for providing such a great 21st century tool and supportive community! All the best!