Hi team, With regards to Call (Live) Translations on VOIP: Is it possible to invoke live translations within the app? (without going into the Call System UI) Is it possible to navigate users from app to Call System UI via an API? (and also invoking the new live translations directly) Will Apple support more languages apart from the current ones? (Currently I see 4 supported languages)

I maintain a couple of CoreImage libraries that provide custom Metal kernel backed CIFilters. In iOS/iPadOS 26, the CIColorKernel.apply() method invoked in the CIFilter subclass fails to add the coreimage::destination parameter to the Metal function call: -[CIColorKernel applyWithExtent:arguments:options:] argument count mismatch for kernel 'FractalNoise3D', expected 13 but saw 12. I've compiled the code with Xcode 26 and deployed to iOS 18 devices without any breakage, so this is definitely an iOS problem, not an Xcode problem. Library here: https://github.com/JoshuaSullivan/SimplexNoiseFilter Feedback ID: FB17874311

Using the official SwiftTranscriptionSampleApp from WWDC 2025, speech transcription takes 14+ seconds from audio input to first result, making it unusable for real-time applications. Environment iOS: 26.0 Beta Xcode: Beta 5 Device: iPhone 16 pro Sample App: Official Apple SwiftTranscriptionSampleApp from WWDC 2025 Configuration Tested Locale: en-US (properly allocated with AssetInventory.allocate(locale:)) and es-ES Setup: All optimizations applied (preheating, high priority, model retention) I started testing in my own app to replace SFSpeech API and include speech detection but after long fights with documentation (this part is quite terrible TBH) I tested the example (https://developer.apple.com/documentation/speech/bringing-advanced-speech-to-text-capabilities-to-your-app) and saw same results. I added some logs to check the specific time: 🎙️ [20:30:41.532] ✅ Analyzer started successfully - ready to receive audio! 🎙️ [20:30:41.532] Listening for transcription results... 🎙️ [20:30:56.342] 🚀 FIRST TRANSCRIPTION RESULT after 14.810s: 'Hello' (isFinal: false) Questions Is this expected performance for iOS 26 Beta, because old SFSpeech is far faster? Are there additional optimization steps for SpeechTranscriber? Should we expect significant performance improvements in later betas?

Hello, I'm trying to receive parquet files using the example that provided in documentation. I've done all required steps but receive constantly error 500 with "Upstream Service Error". By looking into the issues list, seems this error exists for months. Is it possible to get it working?

I was testing audio playback from YouTube in Safari, and the sound was clipping heavily. At first, I thought it might be due to the poor quality of my small sound system. However, when I took a screenshot and the screenshot sound effect itself produced a loud clipping noise, it became clear that this is not a mechanical problem with my speakers, nor an issue specific to YouTube or Safari. This appears to be a system-wide audio issue in macOS Tahoe 26 - Beta 5.

Hello, I'm trying to subscribe to AVPlayerItem status updates using Combine and it's bridge to Swift Concurrency – .values. This is my sample code. struct ContentView: View { @State var player: AVPlayer? @State var loaded = false var body: some View { VStack { if let player { Text("loading status: \(loaded)") Spacer() VideoPlayer(player: player) Button("Load") { Task { let item = AVPlayerItem( url: URL(string: "https://sample-videos.com/video321/mp4/360/big_buck_bunny_360p_5mb.mp4")! ) player.replaceCurrentItem(with: item) let publisher = player.publisher(for: \.status) for await status in publisher.values { print(status.rawValue) if status == .readyToPlay { loaded = true break } } print("we are out") } } } else { Text("No video selected") } } .task { player = AVPlayer() } } } After I click on the "load" button it prints out 0 (as the initial status of .unknown) and nothing after – even when the video is fully loaded. At the same time this works as expected (loading status is set to true): struct ContentView: View { @State var player: AVPlayer? @State var loaded = false @State var cancellable: AnyCancellable? var body: some View { VStack { if let player { Text("loading status: \(loaded)") Spacer() VideoPlayer(player: player) Button("Load") { Task { let item = AVPlayerItem( url: URL(string: "https://sample-videos.com/video321/mp4/360/big_buck_bunny_360p_5mb.mp4")! ) player.replaceCurrentItem(with: item) let stream = AsyncStream { continuation in cancellable = item.publisher(for: \.status) .sink { if $0 == .readyToPlay { continuation.yield($0) continuation.finish() } } } for await _ in stream { loaded = true cancellable?.cancel() cancellable = nil break } } } } else { Text("No video selected") } } .task { player = AVPlayer() } } } Is this a bug or something?

On an iPhone running iOS 26 beta 5, url(for: FilePath("subdir/asset.mov")) most always throws this error: The URL for “subdir/asset.mov” couldn’t be retrieved: “asset.mov” couldn’t be copied to “subdir” because an item with the same name already exists. Yet, contents(at: FilePath("subdir/asset.mov")) always returns Data for a playable AVMovie. How can I avoid this url(for:) error? The asset pack in question is downloaded. The error persists even after pack deletion, redownload, relaunch, and combinations of that. // Assets repo root subdir.aar subdir/asset.mov subdir/asset_thumb.heic subdir/Manifest.json // Manifest.json { "assetPackID": "subdir", "downloadPolicy": { "onDemand": {} }, "fileSelectors": [ { "directory": "subdir", }, ], "platforms": [ "iOS", "visionOS" ] } xcrun ba-package subdir/Manifest.json -o subdir.aar xcrun ba-serve --host 192.168.0.10 -p 443 subdir.aar

I take that MusicKit JS is built with TypeScript, based on the attributions in the script: https://js-cdn.music.apple.com/musickit/v3/musickit.js In the script it points to https://js-cdn.music.apple.com/musickit/v1/acknowledgements.txt – I assume this should be the v3 URL for the v3 version? It returns the same content nonetheless. This contains attributions for TypeScript. Currently there's a third-party effort with DefinitelyTyped, which publishes the NPM package @types/musickit-js. The latest supported version available is v1. However, there is no version compatible with v3. This makes it hard to use MusicKit JS v3 in a TypeScript project. Please publish the types, ideally on the CDN along with the musickit.js file. Also consider publishing an officially Apple supported DefinitelyTyped package, or help to maintain the existing @types/musickit-js to make consuming this even easier.

I am developing an app that plays HLS audio. When using AVPlayerItem with AVURLAsset, can AVAssetResourceLoaderDelegate correctly handle HLS segments? My goal is to use AVAssetResourceLoaderDelegate to add authentication HTTP headers when accessing HLS .m3u8 and .ts files. I can successfully download the files, but playback fails with errors. Specifically, I am observing the following cases: A. AVAssetResourceLoaderDelegate is canceled, and CoreMediaErrorDomain -12881 occurs In NSURLConnectionDataDelegate’s didReceiveResponse method, set contentInformationRequest In didReceiveData, call dataRequest respondWithData resourceLoader didCancelLoadingRequest is called CoreMediaErrorDomain -12881 occurs B. CoreMediaErrorDomain -12881 occurs In NSURLConnectionDataDelegate’s didReceiveResponse method, set contentInformationRequest In connection didReceiveData, buffer all received data until the end In connectionDidFinishLoading, pass the buffered data to respondWithData Call loadingRequest finishLoading CoreMediaErrorDomain -12881 occurs In both cases, dataRequest.requestsAllDataToEndOfResource is YES. For this use case, I am not using AVURLAssetHTTPHeaderFieldsKey because I need to apply the most up-to-date authentication data at the moment each file is accessed. I would appreciate any advice or suggestions you might have. Thank you in advance!

I might have misunderstood the docs, but is Call Translation going to be available for VOIP applications? Eg in an already connected VOIP call, would it be possible for Call Translations to be enabled on an iOS 26 and Apple Intelligence supported device? I have personally tried it and it doesn’t look like it supported VOIP but would love to confirm this. reference: https://developer.apple.com/documentation/callkit/cxsettranslatingcallaction/

After Picture-in-Picture is opened on the camera interface, the custom UI cannot be displayed. Is there any way to make the custom UI visible? If custom UI cannot be displayed, how do teleprompter-type apps in the store manage to show custom teleprompter text within the camera?

ApplicationMusicPlayer is not available on watchOS but all other platforms. Is there a technical reason for that like battery life? Same goes for SystemMusicPlayer and MPMusicPlayerController. I already filed feedbacks for that.

I’m an amateur developer working on a free utility for composers/producers, for which the macOS release needs to create and name RTP-MIDI sessions in Audio MIDI Setup from the command line (so I can ship a small C helper instead of telling users to click through the UI). Here’s what I’ve tried so far, without luck: • Plist hacks: Injecting entries into ~/Library/Audio/MIDI Configurations/*.mcfg works when AMS is closed, but AMS immediately locks and reverts my changes when it’s open. • CoreMIDI C API: I can create virtual ports with MIDISourceCreate, but attempting MIDIObjectGetDataProperty on the apple.midirtp.session plugin always returns err –10836. • Obj-C & Swift: Loading MIDINetworkSession and calling defaultSession, init, setNetworkName: and setting enabled = YES doesn’t produce a new session object in the Network panel. • dlopen/dlsym: I extracted the real CoreMIDI binary out of the dyld shared cache and tried binding _MIDINetworkSessionCreate, _SetName, _SetEnabled, etc., but all the symbols come back null or my tool segfaults. • Plugin registration: I’ve pulled the factory UUID (70C9C5EA-7C65-11D8-B317-000393A34B5A) from /System/Library/Extensions/AppleMIDIRTPDriver.plugin/Contents/Info.plist and called CFPlugInRegisterFactories, but it still never exposes the session-creation calls. At this point I’m convinced I’m either loading the wrong binary or missing one critical step in registering the RTP-MIDI plugin’s private API. Can anyone point me to: The exact path of the dylib or bundle that actually exports the MIDINetworkSessionCreate/MIDINetworkSessionSetName/MIDINetworkSessionSetEnabled symbols? A minimal working snippet (C or Obj-C) that reliably creates and names a Network-MIDI session? Any pointers, sample code, or even ideas about where Apple hides this functionality on macOS 15 would be hugely appreciated. Thanks!

Hello! I'm experiencing an issue with iOS's audio routing system when trying to use Bluetooth headphones for audio output while also recording environmental audio from the built-in microphone. Desired behavior: Play audio through Bluetooth headset (AirPods) Record unprocessed environmental audio from the iPhone's built-in microphone Actual behavior: When explicitly selecting the built-in microphone, iOS reports it's using it (in currentRoute.inputs) However, the actual audio data received is clearly still coming from the AirPods microphone The audio is heavily processed with voice isolation/noise cancellation, removing environmental sounds Environment Details Device: iPhone 12 Pro Max iOS Version: 18.4.1 Hardware: AirPods Audio Framework: AVAudioEngine (also tried AudioQueue) Code Attempted I've tried multiple approaches to force the correct routing: func configureAudioSession() { let session = AVAudioSession.sharedInstance() // Configure to allow Bluetooth output but use built-in mic try? session.setCategory(.playAndRecord, options: [.allowBluetoothA2DP, .defaultToSpeaker]) try? session.setActive(true) // Explicitly select built-in microphone if let inputs = session.availableInputs, let builtInMic = inputs.first(where: { $0.portType == .builtInMic }) { try? session.setPreferredInput(builtInMic) print("Selected input: \(builtInMic.portName)") } // Log the current route let route = session.currentRoute print("Current input: \(route.inputs.first?.portName ?? "None")") // Configure audio engine with native format let inputNode = audioEngine.inputNode let nativeFormat = inputNode.inputFormat(forBus: 0) inputNode.installTap(onBus: 0, bufferSize: 1024, format: nativeFormat) { buffer, time in // Process audio buffer // Despite showing "Built-in Microphone" in route, audio appears to be // coming from AirPods with voice isolation applied - welp! } try? audioEngine.start() } I've also tried various combinations of: Different audio session modes (.default, .measurement, .voiceChat) Different option combinations (with/without .allowBluetooth, .allowBluetoothA2DP) Setting session.setPreferredInput() both before and after activation Diagnostic Observations When AirPods are connected: AVAudioSession.currentRoute.inputs correctly shows "Built-in Microphone" after setPreferredInput() The actual audio data received shows clear signs of AirPods' voice isolation processing Background/environmental sounds are actively filtered out... When recording a test audio played near the phone (not through the app), the recording is nearly silent. Only headset voice goes through. Questions Is there a workaround to force iOS to actually use the built-in microphone while maintaining Bluetooth output? Are there any lower-level configurations that might resolve this issue? Any insights, workarounds, or suggestions would be greatly appreciated. This is blocking a critical feature in my application that requires environmental audio recording while providing audio feedback through headphones 😅

I've just begun to dip my toes into the iOS16 waters. One of the first things that I've attempted is to edit a library playlist using: try await MusicLibrary.shared.edit(targetPlaylist, items: tracksToAdd) Where targetPlaylist is of type MusicItemCollection<MusicKit.Playlist>.Element and tracksToAdd is of type [Track] The targetPlaylist was created, using new iOS16 way, here: let newPlaylist = try await MusicLibrary.shared.createPlaylist(name: name, description: description) tracksToAdd is derived by performing a MusicLibraryRequest on a specific playlist ID, and then doing something like this: if let tracksToAdd = try await playlist.with(.tracks).tracks {    // add tracks to target playlist } My problem is that when I perform attempt the edit, I am faced with a rather sad looking crash. libdispatch.dylib`dispatch_group_leave.cold.1:     0x10b43d62c <+0>:  mov    x8, #0x0     0x10b43d630 <+4>:  stp    x20, x21, [sp, #-0x10]!     0x10b43d634 <+8>:  adrp   x20, 6     0x10b43d638 <+12>: add    x20, x20, #0xfbf          ; "BUG IN CLIENT OF LIBDISPATCH: Unbalanced call to dispatch_group_leave()"     0x10b43d63c <+16>: adrp   x21, 40     0x10b43d640 <+20>: add    x21, x21, #0x260          ; gCRAnnotations     0x10b43d644 <+24>: str    x20, [x21, #0x8]     0x10b43d648 <+28>: str    x8, [x21, #0x38]     0x10b43d64c <+32>: ldp    x20, x21, [sp], #0x10 ->  0x10b43d650 <+36>: brk    #0x1 I assume that I must be doing something wrong, but I frankly have no idea how to troubleshoot this. Any help would be most appreciated. Thanks. @david-apple?

Hello, Has anyone else experienced variations in the accuracy of the playbackTime value? After a few seconds of playback, the reported time adjusts by a fraction of a second, making it difficult to calculate the actual playbackTime of the audio. This can be recreated by playing a song in MusicKit, recording the start time of the audio, playing for at least 10-20 seconds, and then comparing the playbackTime value to one calculated using the start time of the audio. In my experience this jump occurs after about 10 seconds of playback. Any help would be appreciated. Thanks!

Here is the demo from Apple's site This issues is specific to iOS 18. When running this demo, we are getting new text when we have a gap in speaking, the recognitionTask(with:resultHandler:) provides new text which is only spoken after the gap and not the concatenation of old text and the new spoken text.

I'm developing a video capture app using AVFoundation, designed specifically for use on a boat pylon to record slalom water skiing. This setup involves considerable vibration. As you may know, the OIS that Apple began adding to lenses since the iPhone 7 is actually very problematic in high vibration circumstances, ironically creating very shaky video, whereas lenses without OIS produce perfectly stable video. Because of this, up until iPhone 14, the solution for my app was simply to use the Selfie lens, which did not have OIS. Starting with iPhone 14 through iPhone 16 (non-Pro models), technical specs suggest the selfie lens still does not include OIS. However, I’m still seeing the same kind of shaky video behavior I see on OIS-equipped lenses. The one hardware change I see in this camera module is the addition of PDAF (Phase Detection Autofocus), so that is my best guess as to what is causing the unstable video. 1- Does that make any sense - that in high vibration settings, PDAF could create unstable video in the same way that OIS does? Or could it be something else that was changed between the iPhone 13 and 14 Selfie lens? Thinking that the issue was PDAF, I figured that if I enabled my app to set a Manual Focus level, that ought to circumvent PDAF (expecting that if a lens is manually focusing, it can’t also be autofocusing via PDAF). However, even with manual focus locked via AVCaptureDevice in my app, on the Selfie lens of an iPhone 16, the video still comes out very shaky, basically unusable. I also tested with the built-in Apple Camera app (using the press-and-hold to lock focus and exposure) and another 3rd party camera app to lock focus, all with the same results, so it's not that my app just isn't correctly doing manual focus. So I'm stuck with these questions: 2- Does the selfie camera on iPhones 14–16 use PDAF even when focus is set to locked/manual mode? 3- Is there any way in AVFoundation to disable or suppress PDAF during video recording (e.g., a flag, device format setting, or private API)? 4- Is PDAF behavior or suppression documented or controllable via AVCaptureDevice or any related class? 5- If no control of PDAF is available, are there any best practices for stabilizing or smoothing this effect programmatically? Note that I also have set my app to use the most aggressive form of stabilization available, so it defaults to .cinematicExtendedEnhanced, if that’s not available, then .cinematicExtended, etc. On the 16 Selfie lens, it is using .cinematicExtended. As an additional question: 6- Would those be the most appropriate stabilization settings for a high vibration environment, and if not, what would be best?

ImageIO encoding to HEICS fails in macOS 15.5. log writeImageAtIndex:1246: *** CMPhotoCompressionSessionAddImageToSequence: err = kCMPhotoError_UnsupportedOperation [-16994] (codec: 'hvc1') seems to be related with https://github.com/SDWebImage/SDWebImage/issues/3732 affected version iOS 18.4 (sim and device), macOS 15.5 unaffected version iOS 18.3 (sim and device), macOS 15.3

When shooting with an iPhone 15 or later, it’s possible to capture HEIC or JPEG images that include gain map information conforming to the ISO 21496-1 standard. However, during image format transcoding, the HEIC codec is able to preserve the ISO 21496-1 gain map. But when converting from HEIC to JPEG, the gain map is transformed into the Apple Gain Map format instead. Is there any solution to this issue?