Chrome/Chrome OS Accessibility Release Notes: Milestone 75
Laura (Palmaro) Allen
Hi everyone,
Chrome version 75 comes with a number of accessibility fixes and improvements across platforms. Here are some of the notable updates in this milestone.
Highlight
We are releasing our new feature Get Image Descriptions from Google on a rolling basis starting in Chrome version 75! This means that this functionality will roll out to screen reader users on desktop platforms over the summer. Some of you may already see the functionality, and others will get it soon. We’ve included much more detail about this feature below, and are excited to get your feedback.
Windows, Mac, Linux
Chrome 75 makes the incognito window toolbar counter more usable, indicating the number of incognito windows open with a screen reader.
Fixes a bug in Windows where screen readers would incorrectly announce the first line of the Omnibox popup.
Adds an update to fallback on user agent defaults for objects without an explicit role.
Chrome OS
Chrome 75 fixes a bug where when the accessibility on-screen keyboard is enabled, the keyboard would not switch to the full layout.
Now supports dashed accessibility focus ring.
The Accessibility Panel now renders above all other windows, including the virtual keyboard making it easier to navigate.
ChromeVox improvements:
Improved experience for parent access code input fields.
More details here: http://crbug.com/938659, http://crbug.com/938660
Download notifications are now better announced.
Android
Chrome 75 improves contrast for close buttons in tab switcher, and text scaling slider in Accessibility Settings.
Fixes what had been confusing behavior with clickable links in system settings.
Web Platform
Elements using display: contents are now included in the accessibility tree.
Support alternative text for css generated content (with the “Experimental Web Features” Chrome flag on).
Get Image Descriptions from Google - Feature Details
There are millions of graphics and images across the web, many of which provide important context and information to users. However, images are left inaccessible to screen reader users if developers do not add alternative text. Alternative text, often referred to as “alt text” or “alt attributes”, is used within HTML code to describe the function and/or appearance of a given image. Ideally, all developers would add alt text to images, describing the content of the image and any relevant context to screen reader users. But the reality is that a large number of images across the web currently do not have alt text, leading screen readers and/or braille displays to say “image”, “unlabeled graphic”, or a lengthy, often unhelpful verbalization of the image’s file name.
To address some of these gaps, screen reader users can ask Chrome to get automatic image descriptions from Google for images or graphics that are missing alt text. Google's artificial intelligence can attempt to recognize text, objects, landmarks, well-known people, animals, and much more in the images found across the web.
To use this feature, you have a choice of enabling image descriptions for a single use - which describes all the images on the current page once only - or always, which keeps describing images until you turn the feature off.
When using this feature and Google is able to provide an automatic image description, your screen reader will announce either "Appears to be", or "Appears to say" prior to reading the description.
To enable automatic image descriptions for any unlabeled image on a web page:
Open a Chrome browser tab.
Open the context menu.
Most Windows PCs have an applications key on the keyboard, or you can press Shift+F10 to open this menu.
When using the VoiceOver screen reader on Mac, you can press VoiceOver+Shift+ M. Note that by default, the VoiceOver keys are Ctrl+Alt.
On Chrome OS with the ChromeVox screen reader running, you can press ChromeVox+M. Note that by default, the ChromeVox key is the Search key.
From the context menu that opens up, use the arrow keys to find "Get Image Descriptions from Google" and then press the right arrow to open the submenu.
Using the arrow keys, press Space to select either the option to always enable image descriptions, or select the option to enable “just once”.
Alternatively, to always enable automatic image descriptions:
(optional) Sign in to your Chromebook or Chrome browser.
In the toolbar, select More Settings.
At the bottom of the Settings page select Advanced.
Under “Accessibility”, you can choose to “Get image descriptions from Google”.
Note that when you are signed into Chrome, the always setting will persist across devices.
To disable automatic image descriptions:
If you have used image descriptions “just this once”, closing or refreshing the page will disable it.
If you have enabled image descriptions “always”, you can use the context menu option or the settings page as described above.
Keep in mind that Chrome may be unable to generate descriptions for an image, or may choose not to if the image is deemed inappropriate or offensive. In any such cases, the screen reader will announce, “No description available.”
Currently, the full range of image descriptions is only available when using Chrome with the English browser language. To change your browser language, follow these instructions.
Images are sent to Google servers but never saved. Web pages are never modified; for your privacy, a website has no way of knowing this feature is being used. To learn more, review our privacy policy.
Note: Image descriptions are only available to screen reader users through spoken feedback or braille, and do not visually appear on the screen. Since the descriptions are generated using artificial intelligence, there may be cases where the description isn’t fully accurate. If you have feedback, please reach out to our Google Accessibility support team through the channels listed on g.co/disabilitysupport.
| Laura (Palmaro) Allen | | Google | | Chrome & Chrome OS Accessibility Program Manager | Pronouns: she/her |