Forum Discussion
Storyline 360 Update 99
On Tuesday, we shipped Update 99 of Storyline 360. This update has tons of fixes and enhancements, including an optimization to how closed captions are loaded by the player, the ability to drag and drop media files directly into the Media Library, and a fix for an issue that kept the NVDA screen reader from properly following focus order. The part of this release that I’m most excited about, however, is the addition of AI generated alt text for images.
In Update 99, authors have the ability to ask the AI Assistant to create alt text for any image in their course. This can be done using new buttons in either the Size and Position dialog, or the Media Library side panel. In the example here, I’m dragging three images into the Media Library depicting a rock, some paper, and scissors. For each image I’m using the “Generate alt text” button to have AI describe the contents of the image. Here’s what it came up with for each:
Rock: A shiny black coal-like mineral with uneven texture and jagged edges.
Paper: A blank spiral-bound notebook on a wooden stand, flanked by two small potted plants on either side.
Scissors: Two small ornate scissors with gold handles, one styled like a bird and the other with a traditional design, placed on a light background.
It’s important to note that human review of the AI generated alt text is still very important. AI might fail to convey the context that you want for learners using assistive technology like a screen reader or simply misinterpret what the image contains. We will be capturing analytics and feedback from customers about the overall quality of the generated alt text so we can continue to refine and improve in the future.
Have you used the AI Assistant to generate alt text in your courses yet? If so, I'd love to hear about how it worked for you in the comments!
If you'd like to learn more about this feature or any of the other things that shipped with Update 99 check out the resources below:
The 100th update of Storyline 360 is coming up on May 27th and we have some very exciting new features slated for release! If you want to get a sneak preview, some of those features are already available in the Storyline private beta program. If you’d like to participate, simply email beta@articulate.com to get started!
8 Replies
This is a good step but there should be a setting that lets me tell Storyline that whenever I add an image to a slide, the AI should automatically add ALT text to it.
The tedium of making Storyline courses WCAG compliant must be reduced by a factor of...well, I don't know the exact amount, but it needs to come down by A LOT in order for it to really be practical to implement all of the necessary accommodations. That means automating everything that can possibly be automated. Automatic insertion of ALT text seems like "low-hanging fruit" in this regard. Don't make me authorize AI-generated ALT text for every individual image. Just let the AI take care of this whenever I add an image to a slide.
NOTE: this philosophy also applies to audio and video. Whenever I add an audio file to a slide, an AI process should kick off that automatically generates closed captions for me. Whenever I add a video to a slide, it likewise should automatically get closed-captioned. Additionally, it should get an AI-voiced audio description of its contents.
You can take this automation further still and I urge you to do so. For example, when I create any object off-slide (e.g., in the margin to the left or right of the slide) the Object is visible to accessibility tools should default to unchecked to mimic the fact that such objects would not be visible to sighted learners in the published output. Likewise, if I drag this object onto the slide, Storyline should automatically switch the object's visibility to accessibility tools to visible to match that objects on the slide are visible to sighted learners in the published output.
It's great that Storyline gives me the ability to edit ALT tags, closed captions, and object visibility to accessibility tools--but that is only the bare minimum. What is really needed is to automate as many of the accessibility and localization decisions as possible so that I only have to edit these things if I don't like what the AI generated. That would save a huge amount of time and tedium, and it seems you have most of the core technology already implemented in Storyline now, so please take a hard look at how much you can automate for us.
RayCole-2d64185 Thank you for taking the time to leave this feedback. I agree that there is a lot more we can do to ease the creation of WCAG compliant courses, and we're only getting started. We want to be very deliberate and iterative with our approach here, ensuring that the AI we add is truly valuable and useful to authors before taking measures like having it automatically create alt text for all images. We also feel that it's important for authors to review all AI generated accessibility content for accuracy and to ensure it fits the needs of that particular course.
The Accessibility Checker tool we plan to release soon has been designed to make it easier than ever to identify and resolve a11y issues in your course. The initial version of this tool won't solve every problem you have, but we will be listening to feedback and continuing to evolve it over time to make it as useful as possible.
As for the "Object is visible to accessibility tools" option: you're completely correct. We need to improve how this works and are considering some options.
Thank you again for the feedback, I truly appreciate it.
Thanks for the reply! I am eagerly-awaiting the accessibility checker.
Some really great ideas there, particularly the 'object off-slide' idea - brilliant.
My personal preference would be a toggle for the "Don't make me authorize AI-generated ALT text for every individual image" feature mentioned. 99% of images used in my modules are purely decorative and I wish there was a setting whereby they could automatically be added with the Accessibility toggled off by default, so I could then manually (or with AI) add Alt text when required.
Also, it would be great if the Focus Order dialog box had a "Reverse Order" button because I do not know why so many items are listed in an 'ascending' manner when I intended things to be tabbed to in a descending order. It would save so much time, just like the ability to now drag a bunch of items instead of individually.
I'd like to see the alt text generator be used to simply caption an image. In Rise I think most content images should have a short description of what they are, the AI tool would be great for this. (of course images need the ability to have caption text underneath which is not always the case e.g. tabs/accordions)
The AI alt tag generator in Storyline is far superior to the one in Word, much more meaningful text but Word does have one advantage of an option to autogenerate for all images in a document. This is needed in Storyline and Rise as its too slow to go into each image separately (but at least the Media library has this feature).
- The alt text generator doesn't seem to work on videos.
- Also is it too detailed when being used on plain buttons: "A rectangular button with the word "Start" centered in black text on a white background." Wouldn't a screen reader, read 'Button. A rectangular button...'
- And on icons its unreliable...
Thanks so much for your feedback on the AI-generated alt text feature—we really appreciate you taking the time to share your experience.
You’re right that the AI currently doesn’t provide ideal results for videos. It generates alt text based on the first frame (or thumbnail) of the video, which can miss the broader context or content. We’re keeping an eye on feedback like yours, and if we continue to hear concerns, we may explore limiting AI-generated alt text for certain media types like video, where it may not add meaningful value.
You also bring up a great point about buttons. For simple, functional elements like “Start,” "Submit,” "Previous,” "Next,” and other interactive buttons, the generated alt text can be overly verbose and possibly redundant when read by screen readers. We also recommend not using AI alt text for these types of buttons. However, it’s a helpful reminder that even with AI support, author review is essential to ensure clarity and avoid repetitive or unnecessary details.
Regarding icons, we've found that the quality of AI-generated alt text can vary depending on how visually distinct the icon is. In my testing, I've generally found that the AI does pretty well with the icons I've tested with. We're continuing to fine-tune the model and its outputs to improve accuracy, but again—manual review is key to ensuring accessible and meaningful content.
The user guide for this feature was recently updated with additional tips that might also be helpful.
Thanks again for the thoughtful insights. We’ll continue improving this feature based on feedback like yours. If you have any additional examples or suggestions, feel free to share—I'm listening!
Thanks,
Ronnie Pilman, CPACC
Senior QA Engineer | Accessibility Lead
