I have a UITextField in my application for entering a state. If I tap on it, a UIPickerView pops up and let's the user select a state (but they can still type too). The issue relates to Full Keyboard Access. If we select the UITextField using an external keyboard, the UIPickerView appears, but in order to get to it the user has to tab through the whole view controller to get to the UIPickerView at the end. What would be nice is to a) move focus directly to the UIPickerView (have it highlighted in blue and scrollable right away with keyboard) or b) make the UIPickerView the next view that's accessible when tabbing over or using the arrow keys. I've tried using: UIAccessibility notifications (both .screenChanged and .layoutChanged, with and without a delay). This ended up only announcing the view, but didn't help with full keyboard access. Making the UIPickerView a first responder when it appears. Attempting to change the accessibilityElements order (but with so many views and views within views, this isn't really a viable option either). Pressing tab + -> (tab and right arrow button) will quickly take the user to the end of the chain of accessibility elements, in other words, to the UIPickerView. But there has to be a cleaner way of just automatically setting the focus to the UIPickerView or making it the next element by pressing the arrow key.

I have some doubts about how VoiceOver handles focus when the screen updates. When a new UIViewController is pushed onto a UINavigationController or presented modally, how does VoiceOver decide which element to focus on? Is there a way to control or customize this behavior? In a UISplitViewController, when an item is selected in the primary view controller, the focus should shift to the relevant content in the secondary view controller. How can we ensure that VoiceOver correctly moves focus to the right element in the secondary panel?

Hi everyone, I’ve been analyzing the current state of Sign Language accessibility tools, and I noticed a significant gap in learning tools: we lack real-time feedback for students (e.g., "Is my hand position correct?"). Most current solutions rely on 2D video processing, which struggles with depth perception and occlusion (hand-over-hand or hand-over-face gestures), which are critical in Sign Language grammar. I'd like to propose/discuss an architecture leveraging the current LiDAR + Neural Engine capabilities found in iPhone devices to solve this. The Concept: Skeleton-based Normalization Instead of training ML models on raw video frames (which introduces noise from lighting, skin tone, and clothing), we could use ARKit's Body Tracking to abstract the input. Capture: Use ARKit/LiDAR to track the user's upper body and hand joints in 3D space. Data Normalization: Extract only the vector coordinates (X, Y, Z of joints). This creates a "clean" dataset, effectively normalizing the user regardless of physical appearance. Comparison: Feed these vectors into a CoreML model trained on "Reference Skeletons" (recorded by native signers). Feedback Loop: The app calculates the geometric distance between the user's pose and the reference pose to provide specific correction (e.g., "Raise your elbow 10 degrees"). Why this approach? Solves Occlusion: LiDAR handles depth much better than standard RGB cameras when hands cross the body. Privacy: We are processing coordinates, not video streams. Efficiency: Comparing vector sequences is computationally cheaper than video analysis, preserving battery life. Has anyone experimented with using ARKit Body Anchors specifically for comparing complex gesture sequences against a stored "correct" database? I believe this "Skeleton First" approach is the key to scalable Sign Language education apps. Looking forward to hearing your thoughts.

Request: Name Recognition → Shortcut for SOS Flashlight + Vibration Right now, iOS Name Recognition works, but all I can do is flash the tiny notification light. It would be much more useful if Name Recognition could trigger a Shortcut. That way, I could set it to flash the flashlight in an SOS pattern and vibrate, making the alert impossible to miss. I tried using Custom Alarm, but it won’t let me record my spoken name, so it doesn’t really solve the problem. If Apple allowed Name Recognition to trigger Shortcuts — or expanded “Custom” to support names/words — this would open up far more practical, real-world alerts.

I have a couple follow up questions after the "Accessibility technologies group lab". I know it was briefly mentioned that user feedback is an excellent way to grow inclusivity in the design an app and utilizing these forums were one for example. Is inviting folks here on the forum via test flight a reasonable approach to this for a solo developer? Are there other strategies, avenues, or examples to promote user feedback?

I have a view dynamically overlaid on a UITableView with proper padding (added when certain conditions are met). When VoiceOver focuses on a cell beneath this overlay, the focused element does not scroll into view. I’ve noticed similar behavior in Apple’s first-party Podcasts app. Please find the attached image for reference. How can I resolve this issue and ensure VoiceOver scrolls the focused cell into view?

I’m developing an ARKit application where I aim to attach procedurally generated audio to detected planes in the environment. While using a static audio file with SCNAudioSource and SCNAudioPlayer works as expected, integrating procedural audio via AVAudioSourceNode does not produce any sound, nor does it generate any error messages: Stack Overflow Post Working Implementation with Static Audio File: let audioPlayer = SCNAudioPlayer(source: audioSource) node.addAudioPlayer(audioPlayer) Attempted Implementation with Procedural Audio: // Audio generation code } let audioPlayer = SCNAudioPlayer(avAudioNode: audioNode) node.addAudioPlayer(audioPlayer) In this setup, the AVAudioSourceNode successfully generates audio when connected directly to an AVAudioEngine. However, when used with SCNAudioPlayer and attached to an SCNNode, it fails to produce sound. What doesn’t work is creating some procedural audio with an AVAudioNode, as documented here: Apple docs Additionally, I explored the WWDC18 AR game project, SwiftShot, which utilizes SCNAudioPlayer(avAudioNode:). After updating it for the latest Xcode, the graphics function correctly, but the audio does not play. I also noted that the Apple documentation mentions an audioPlayerWithAVAudioNode: method, stating: Using this initializer is typically not necessary. Instead, call the audioPlayerWithAVAudioNode: method, which returns a cached audio player object if one for the specified AVAudioNode object has already been created and is available for use. However, this method does not appear to be available in Swift. Any insights or guidance on this matter would be greatly appreciated.

When I am doing a file search, in TextEdit, and on certain webistes the space bar will quit functioning as soon as i start typing. If I hold down the "Option" key it allows the space bar to work as normal. I have checked every setting I can think of and nothing has helped.

Hello, When I listen to title in my app with VoiceOver, it makes a strange sound. This characters make with Korean+number+Alphabet. Is this combination makes some strange sound with voice over? I would like to ask if Apple can fix this issue. Thank you.

Amogst the two languages my app would have lets say 10% and 90%. I am launching an app for a unlisted Primary Language. I consider whatever is inside the app as the primary and that wont be English. The secondary language

I have an HTML select that has Spanish text in the options. When VoiceOver reads the selected option (unopened), it switches to Spanish as expected. However, when you open the select box and browse through the options, it uses the English voice to read the Spanish text. I have tried adding lang on to the select tag and the option tag but neither helps https://codepen.io/grahamfowles/pen/VYYRxMK

Hi everyone, I’m developing a React Native iOS app that includes a custom keyboard extension for sending stickers across apps. The project builds successfully, and the main app installs fine on my test device. However, I’m not seeing the keyboard extension appear under Settings → General → Keyboard → Keyboards → Add New Keyboard, which means I can’t activate it or grant access. At this point, I’m not even sure if the extension is actually being installed on the device along with the main app. Here’s what I’ve done so far. I created a Keyboard Extension target in Xcode, set the correct bundle identifiers and provisioning profiles, and enabled “Requests Open Access” in the extension’s Info.plist. I built and installed the app on a physical device rather than the simulator to ensure proper testing. My main questions are: how can I confirm that the extension is being installed on the device, and if it isn’t, what might prevent it from installing even though the build completes successfully? Any insights, troubleshooting steps, or guidance would be greatly appreciated.

I am seeing a strange issue where NSObject accessibilityRespondsToUserInteraction returns true on Simulator but false on device. Checking the same object on simulator with Accessibility inspector I see the object traits as image so why would it return true in that case? Are there any other way to check the the item might be accessibilityRespondsToUserInteraction OR Clickable beside that property and traits? (Or is it just another bug)

Context: We are using UIKit to provide accessibility in our app for our iOS users. Our app majorly contains documents/books that user can read. Issue: The issue is VoiceOver is skipping the lines given to it when there are some leading spaces in it. We have observed this issue in different languages. This is only happening for line granularity, other granularities seems to be working as expected. Implementation: We are using below API's to provide line content to voice over. UIAccessibilityReadingContent - accessibilityPageContent - accessibilityFrameForLineNumber - accessibilityContentForLineNumber We are creating UIAccessibilityElement objects to pass to VoiceOver and each UIAccessibilityElement implements UIAccessibilityReadingContent to provide readable content. We also use below APIs to cross element boundaries for all granular navigations. accessibilityNextTextNavigationElement accessibilityPreviousTextNavigationElement We want to know whether skipping the line when provided with leading spaces is expected or a bug in UIKit.

In SwiftUI, the date picker component is breaking in colour contrast accessibility. Below code has been use to create date picker: struct ContentView: View { @State private var date = Date() @State private var selectedDate: Date = .init() var body: some View { let min = Calendar.current.date(byAdding: .day, value: 14, to: Date()) ?? Date() let max = Calendar.current.date(byAdding: .year, value: 4, to: Date()) ?? Date() DatePicker( "Start Date", selection: $date, in: min ... max, displayedComponents: [.date] ) .datePickerStyle(.graphical) .frame(alignment: .topLeading) .onAppear { selectedDate = Calendar.current.date(byAdding: .day, value: 14, to: Date()) ?? Date() } } } #Preview { ContentView() } attaching the screenshot of failure accessibility.

Hi everyone, My team and I are developing an accessibility-focused VisionOS app (MindTap) as part of a university project, aiming to support individuals with Locked-In Syndrome using Brain-Computer Interface (BCI) signals to trigger interactions (e.g., tapping) within the Apple Vision Pro environment. Problem 1: Simulating Eye Tracking in Simulator We are testing onHover with Send pointer to the device under I/O > Input in the simulator, and while it mostly works (a bit laggy), we found that onHover won't function on the actual Vision Pro hardware. From what I understand, we should be using FocusState for proper gaze interaction, but testing this requires the physical device. Is there any workaround or official Apple-recommended way to simulate Focus-based gaze detection without a real Vision Pro? Problem 2: WebSocket-triggered "Click" doesn't work outside the app We successfully use WebSocket to send a custom signal (a "1" from the brain signal device) to trigger an action inside our app. However, when the user opens a third-party app like Apple News, the WebSocket-triggered "click" no longer works. We suspect this is due to sandbox restrictions or lack of system-level permissions. Is it possible in anyway to: Trigger interaction events outside the app using custom input (like BCI via Websocket)? Access system-wide click/tap simulation APIs from within VisionOS apps Integrate this with accessibility services (like Voice Control or AssistiveTouch) We'd appreciate any official guidance or tips from others building similar accessibility apps with alternative input methods in VisionOS. Thanks in advance for any insight you can provide!

On iOS, there is accessibilityLanguage.

I'm developing a calculator app and working to ensure a great experience for both VoiceOver and Braille display users. For expressions like (2+3)×5, I need two different accessibility outputs: VoiceOver (spoken): A descriptive string like “left paren two plus three right paren times five,” provided via .accessibilityValue. I'm using a custom spellOut function since VoiceOver doesn't announce parentheses—which are kind of important when doing math! Braille (symbolic): The literal math string (2+3)×5, provided using .accessibilityCustomContent("", ...), with an empty label so it’s not spoken aloud. The issue: I don’t have access to a Braille display device and Xcode’s Accessibility Inspector doesn’t seem to show the custom content. Is there any way to confirm that custom Braille content is being set correctly in Simulator or with other tools? Or…is there a "math mode" in VoiceOver that forces it to announce parentheses? Any advice or workarounds would be much appreciated! Thanks, Uhl

Triple tap for screenshot->notification->triple tap detected becomes a part of the screenshot and obscures the top part of screenshot. Thanks

We have an electron app developed for Mac. We would like to restore the user data previously saved in downloads once user installs the app from store and first launch. But MAS has restrictions with ""com.apple.security.files.downloads.read-write". We have enabled the user access in Entitlement files and request user permission before access What options can be user to auto restore the data from downlodas?