What's new in Accessibility in iOS 27 for Blind and DeafBlind Users

Introduction

15 years ago, I produced my first "What's New in iOS Accessibility" article. Every September since, I've been documenting what my testing of the betas has yielded. There are many welcome mainstream changes in this year's iOS 27, such as the upgraded Siri app, new parental controls, a slider to adjust Liquid Glass, further enhanced Apple Intelligence features, and much more. The new Siri has its own app, and coverage of it is beyond the scope of this article other than to mention it's much more versatile and can assist with all kinds of tasks from asking about your personal data to recipes and so on. Lots of individuals will be reviewing these new features. As I've done for many years now, I'm documenting my findings after testing the new version of iOS throughout the summer to discuss accessibility-specific features which impact blind and DeafBlind users.

Following what has become a tradition, Apple began discussing enhancements coming to accessibility with its announcement ahead of Global Accessibility Awareness Day. More info about forthcoming features was detailed during the Worldwide Developers Conference Keynote. Devices that will work with iOS 27, according to Apple, are the iPhone SE 2 and later, as well as the iPhone 11 and later. The page linked to above is also a good resource for mainstream changes coming in iOS 27.

VoiceOver

Though there are fewer changes in iOS 27 than what were found in iOS 26, VoiceOver has gained quite a bit of new functionality. Whether you are a speech user or a braille one, there are a lot of new offerings in this year's release.

Pronounce It Together

There is now a setting which allows a VoiceOver user to import or share their pronunciation dictionary. This feature could be helpful when, for example, text-to-speech engines do not pronounce things correctly in local dialects. Instead of everyone having to make the changes, one person could compile it and share it with other users in a similar situation. It could also be used as a form of prank. Imagine changing the pronunciation of a friend's name to a friendly insult along with other changes and they later encounter it. These settings can be found under Settings> Accessibility> VoiceOver> Speech> Pronunciations. Selecting 'Import' will bring up the Files function and allow the user to browse to the file they wish to import. Sharing one's own pronunciations brings up the standard Share Sheet, where the file can be sent to any number of places or contacts.

Hide Your Cursor

For those who may not be aware, the VoiceOver cursor appears as a black rectangular outline surrounding the item that currently has VoiceOver focus. While handy in some situations such as showing a sighted user where VoiceOver focus is set, helping a low vision user who tracks VoiceOver focus, or when demonstrating VoiceOver support to a developer, there are certainly reasons individuals may wish to disable it. This is because it can also serve as a distraction when the VoiceOver cursor itself isn't an important part of the iPhone's purpose. For example, when using the iPhone for face-to-face communication as a DeafBlind user, it can distract the person you are attempting to communicate with from the text you are typing. Some may wish to hide the visual VoiceOver cursor during screen sharing. To adjust this setting, head to Settings> Accessibility> VoiceOver> Cursor Visibility, and choose from showing the cursor, hiding it always, or hiding it during screen sharing.

Navigating in New Ways with Explore By Touch

iOS 27 brings new options to precisely control VoiceOver's reading of text when using Explore by Touch, or, as it's more commonly referred to, 'dragging your finger around the screen'. Options for Text Exploration are now Paragraph, Line, or Sentence. For example, when set to 'Sentence', VoiceOver will only speak the sentence under your finger instead of the entire element as it previously did. Users can adjust this by going to Settings> Accessibility> VoiceOver> Text Exploration. Note that changes made in Text Exploration only apply to Explore by Touch, and will not impact swiping through content. At least, not this particular setting.

It's for Swipers Too

A separate setting, called Text Navigation, offers the ability to modify how VoiceOver moves when swiping left and right through the screen with one finger. The available options are Line, Sentence, Paragraph, and Element. In addition to allowing one to more easily read blocks of text that VoiceOver sees as one large element, having these two new sets of options can give VoiceOver users other choices than the rotor for text navigation. To check out the new settings for swiping, head to Settings> Accessibility> VoiceOver> Text Navigation. Remember that you will need to adjust settings for Text Exploration and Text Navigation separately.

Hints Become Customizable

For a lot of the clients I work with who use speech, the hints offered by VoiceOver can be helpful, but there comes a point where they cause interference with touchscreen navigation once a user advances in their journey of learning VoiceOver. For example, "Double Tap to Open" can grind on one's nerves after a while. When I first started learning to use an iPhone with iOS 4, this was also something I became frustrated with. If the system hints become a nuisance, the user now has more control over them.

iOS 27 brings the ability to adjust the amount of delay between when VoiceOver provides active information about what is in focus and when the hint will be conveyed. To configure this, head to Settings> Accessibility> VoiceOver> Verbosity> Hints, and then either type the duration in seconds ranging from 0.05 to 2.0, or use the 'Decrement' and 'Increment' buttons. The default is 0.8 seconds. It's also now possible to specify if users want all hints, app-specific hints, or no hints at all. You can find these options on the same screen.

No Modifier Required for These Functions

For VoiceOver commands to be assigned to keyboard shortcuts, Bluetooth keyboard users have always had to include their chosen VoiceOver modifier along with the command. This is no longer necessary with the function keys located in the top row of many Bluetooth keyboards. The setting must be enabled, which one can do by going to Settings>Accessibility>VoiceOver>Commands and setting the Switch Button for "No Modifier Keys" to on. Instead of having to press, for example, VoiceOver Modifier f8 to launch the VoiceOver Settings menu, users can now simply press F8.

Live and Open for Questions

Live Recognition was first integrated into VoiceOver with iOS 18. iOS 27, coupled with Apple Intelligence, gives Live Recognition even more power. Activating the 'Ask' button will allow the user to ask about what is currently in the camera's view. It's also possible to tell Live Recognition to focus on a specific thing in your environment. As long as that item is able to be tracked, its actions and movements will often be reported by VoiceOver.

Ask is a feature that is highly customizable. Going to Settings> Accessibility> VoiceOver> Live Recognition allows the user to set up some other options. One can change the default question asked when the button is activated, make the default input either dictation or typing, and have the Volume buttons take a still photo for a follow-up or more detailed description. Ask can be set up with a gesture, Bluetooth keyboard command, the Accessibility Shortcut, or a Perkins-style keyboard command. If none of those are what the user wants, there is also the ability to launch the Ask feature using the Action Button. It can be enabled by going to Settings> Action Button, selecting the Accessibility category, and then choosing 'Ask a Question'.

The View From iOS

In my view, pardon the pun, the largest upgrade for VoiceOver and braille users in iOS 27 is the massive overhaul of the Image Recognition feature. This is thanks to Apple Intelligence, but I'm sure those using iPhone 15 and earlier devices aren't feeling very thankful, since they do not have access to these features. Unlike the multi-step processes involved with using third-party services for detailed descriptions, Image Recognition descriptions typically take less than three seconds. They can be generated on virtually all images the iPhone encounters, and the user does not need to download or upload each image they wish to have described. It is also possible to describe the screen of the device itself, to ask questions about the image in focus, and to generate a description with the ability to ask a follow-up question immediately. These features can be activated through the VoiceOver rotor or by adding specific functions to gestures. The two devices I have upgraded at this point had the Image Recognition options in the Rotor. If, by chance, your Rotor does not have the Image Recognition option after upgrading, it can be added by going to Settings> Accessibility> VoiceOver> Rotor> Rotor Items and selecting Image Recognition. Since there are up to seven different options for Image Recognition, Apple has also given users the ability to add and remove what they wish to this rotor item. It can be configured by going to Settings> Accessibility> VoiceOver> Rotor> Image Recognition Rotor. Each of the available options will be discussed below.

Image/Item Description will allow the user to get a description of the image or item that VoiceOver focus is set to. Based on my testing, it is generated in less than three seconds. It will be delivered in the form of a VoiceOver announcement. For those using braille, one can read the description at their own pace by pressing Space with N. When done, pressing space with n will return the user to where they were before. I find this particular function to be wonderful for generating descriptions of things in my Facebook feed, my growing photo library, album art in the Music app, and when reading photo-heavy articles on various news sources. Interestingly, this function is the only one that can be mapped to a Perkins-style keyboard. The other options discussed below are not available for users of Perkins-style keyboards. Using the Copied Speech Rotor option launched with iOS 26, users can copy the description to wherever they want on their iOS device. A simpler solution in some situations is available as discussed below.

Ask About Image/Item is another option, which does exactly what the choice says. After selecting this, users are presented with a text field which allows them to type a question. Pressing return or enter will generate a response. The nice thing is that you do not need to generate a general description before asking a question. Placing VoiceOver focus on the image is all that is required. With many third-party applications, one must first wait through getting the general description before they can ask a question. I have solved simple CAPTCHAs by asking my iPhone to provide me with only the text in the image. I can then copy it and solve CAPTCHAs. I have not tried more complex puzzles.

Explore Image/Item allows you to get the same description as you get when generating an Intelligent Description, but it is static text instead of a VoiceOver announcement. After the text description, there is an option to ask a question either by typing or voice. Also, when focus is on the description, one can flick up with one finger, or use whatever alternative method, to copy the description.

Ask About Screen allows the user to get a description of the screen on their iOS device. It functions the same as the Ask About Image/Item choice, but covers the entire screen. I've found this feature handy for understanding what buttons without labels in applications do. This is far more preferable to activating the button to see what happens.

The other two options are Describe Screen and Explore Screen. These work the same as the others documented above, only this applies to your entire screen instead of one item. I mentioned browsing photos above, and when in the Photos app, there is yet another rotor option available called Save Intelligent Image Description. When selected, this will add the description generated to the photo itself which will then be available later.

A final setting is the ability to show a list of the recognition options discussed above. It does not have a touch screen gesture assigned to it, but the user can create one. On a Bluetooth keyboard, pressing VoiceOver Modifier, Shift, and R will bring up this menu. Sorry, braille display users, there appears to be no Perkins-style keyboard option.

Automatic Image Descriptions

Users can head to Settings>Accessibility>VoiceOver>VoiceOver Recognition>Image Descriptions to configure whether image descriptions are provided automatically. One can configure whether they want automatic image descriptions or whether they are conveyed as a fallback when images have alt text. Note that this setting applies to the classic image descriptions, but one will not receive a detailed description such as the ones provided by Apple Intelligence.

Braille

There are some changes to braille outlined below. As a general usage note, writing longer passages of text seems to be a bit more reliable than it was in all of the iOS 26 releases. Further, there were some issues in iOS 26 with cursor routing buttons not routing correctly and focus jumping back to the top in longer blocks of text while the user was reading. Based on my use, all of these issues have been improved significantly.

Predict More Predictably with Braille Screen Input

Braille Screen Input now uses the same Predictive Text gestures as the default onscreen keyboard. When typing on the standard touchscreen keyboard, iOS uses Autocorrect and Predictive Text to identify the word you probably intended based on the letters entered, nearby keys, sentence context, and commonly-used language. Braille Screen Input now uses this same prediction system: after entering a word, users can swipe up or down with one finger to move through suggested words or corrections, then continue typing once the desired suggestion is selected.

Further, it is now possible to type part of an emoji name and swipe up or down with one finger to find it. Functionality of the BSI Autocorrect requires the primary software keyboard Autocorrect setting to be enabled in Settings> General> Keyboards.

Finally, iOS 27 adds the ability to more easily insert alternative characters. To activate alternative character selection, hold the dots of the character you wish to insert until you hear a sound, then release and swipe up or down with one finger to select the desired symbol.

A More Flexible Cursor

Shifting to braille display users, there is a new setting called 'Snap To Print Text Cursor', which can be found by going to Settings> Accessibility> VoiceOver> Braille> Cursor Routing. The options are the new Snap To Text Cursor or Any Braille Cell. By default, Snap To Print Text Cursor was selected when I upgraded. The 'Snap to Print Text' setting places the cursor on the closest print text equivalent even when contracted braille is enabled. This setting is particularly helpful for those braille display users using QWERTY keyboards to type while also using contracted braille for reading, as it removes some ambiguity from where the cursor is located.

Text Field Labels That Stick

Prior to iOS 27, when activating a text field, the text label would disappear entirely. In iOS 27, the label persists even when the text field is active. The label itself appears between two indicator symbols which clearly distinguish the label from the text entered. It can be configured by going to Settings> Accessibility> VoiceOver> Braille> Text Field Labels. The options are to show an abbreviated version of the label or the full thing. Selecting 'Collapse' will provide two indicators without the text label. Selecting 'Full' will display the entire text field label. You can collapse the label by pressing a cursor routing button. If you disable everything in this menu, the behavior will be as it always was in past versions of iOS.

Braille Access Expands File Support

Braille Access now supports plain text and PDF files. As such, the formerly named 'BRF Files' is now simply called 'Files'. To import these file types, users can follow the same procedures they already use to get BRF content imported. Files are still stored in the 'BRF files' folder on iCloud, so any method a user would utilize to copy text and PDF content is the same.

I found that opening plain text files worked quickly, and it was nice to have them imported. Unlike many applications for braille readers, Braille Access tends to remember the place I was in a file when I open it again. PDF content will display, but seems to depend on how well the creator of the file treated accessibility. Formatting such as headings is not preserved. There also appears to be a file size limit, as I was not able to load a PDF file that was 15 MB due to the images it contained.

New Features Bring New Keyboard Shortcuts

Apple has added a few new options for Perkins-style keyboard shortcuts that people may take advantage of. Under the VoiceOver category, one has been added for launching the Ask function found in VoiceOver Recognition, while another was added for generating an Intelligent Description with Image Recognition. There are no new commands for the other Image Recognition options discussed above. Another mapping option was made available for ignoring trackpads.

Hearing

On-demand Sign Language Interpreters on FaceTime

As part of Apple's Global Accessibility Awareness Day announcement in May, the company shared preliminary details of an API that providers of sign language interpretation apps could use to allow callers to add an interpreter to their FaceTime call. As of the time of writing, there do not appear to be any announcements concerning providers utilizing this API, but hopefully that will come soon.

Auto-Caption This Video

Captioning automatically produces synchronized subtitles from spoken audio when a video has no existing captions. Apple says it works with videos recorded on the iPhone, videos sent to the user, streamed videos, and other uncaptioned videos played through supported system playback interfaces. When I enabled captioning in the Photos app while playing a video, there were several visual options available for adjusting the output. However, I was not able to find the generated captions with braille. Thankfully, Live Captions under Braille Access works well.

Conclusion

iOS 27 may not bring as many accessibility changes as its predecessor, but several of its additions are substantial. The expanded image-recognition tools can give blind users faster and more independent access to photos, unfamiliar controls, and other visual information, while the new navigation and customization options allow VoiceOver users to better tailor the experience to their needs. Braille users also gain welcome improvements, including more reliable text editing, clearer text-field labels, and support for additional file formats in Braille Access. Some of the most significant new features require Apple Intelligence, leaving users of older compatible devices without access to everything discussed in this article. There are also a few places where braille keyboard support remains incomplete. Even with those limitations, iOS 27 offers a useful collection of accessibility improvements. As always, individual experiences will vary, and readers are encouraged to share their own observations and bugs in the comments. I also encourage users to check out the AppleVis list of bugs and fixes. If needed, more information on how to update the software on your device is available on this Apple Support page.

T