Hi, Safari supports playing back webm files (such as those recorded by Chrome/Firefox browsers with the MediaRecorder API), but our native app is unable to play these back with AVPlayer. Is there a supported way to do this? Right now we have to use a WebView for this! Would also love to ExtAudioFileURLOpen or process these files with AVAudioEngine. Have I missed something obvious here?

I am running into zippering and zipper storms in making my Auv3 synth. Do you have advice? It is a big synthesizer, big cinematic sounds, big presets, dual mode, I am mostly hitting zipper storms when trying to play fast on big preset recipes. Is there something I need to know to avoid zippering?

Are there plans to extend the session or MediaDevice capabilities to be able to add and remove speakers via the Lock Screen and Control Center? For example, user may have multiple speakers, they start playback on a subset of them and would like to add other speakers or remove some from the session through Control Center.

Hi everyone! Adam here, the developer of Albums. Bummed not to get to have our yearly WebEx reunion where I beg y’all to add an endpoint to the Apple Music API to allow deleting items from a user’s library. But that’s not what this thread is about. Back in October, thanks to your team’s multi-year efforts, I was finally able to ship Albums for macOS. It’s been a genuine dream come true to use it on the platform I always envisioned it on, and I hear from users all the time who feel the same way. I know it was a significant engineering effort, and I’m genuinely very grateful. I’ll be judicious with my time here (I’d love to chat in more detail about my adventures in MusicKit on the Mac somehow, sometime). These are the main things on my wishlist for MusicKit on the Mac. There are three main things I hear from users that my app “can’t do” that can be dealbreakers for them. The first two are AirPlay support and a volume slider. Users are only able to AirPlay using the control center utility, because AVRoutePickerView does not work with ApplicationMusicPlayer (FB13934910). Relatedly, the MPVolumeView does not work with ApplicationMusicPlayer (FB21042385), so I can’t allow users app-specific volume control for my app. The lack of those two things is a real detriment to my app being able to be taken seriously alongside all of the other music player apps on the platform for which those features are table stakes. I know there’s a challenge here given the playback actually happens in the subprocess, but hoping some progress can be made here. The other issue is that library tracks played in ApplicationMusicPlayer do not update the last played date or play count in Music.app or in the user’s iCloud Music Library (FB17675148). Some people refuse to use the app for that reason, and I can’t say I blame them. I’ve only been able to test this briefly in Golden Gate, but it seems like this is still the case. Are you able to share anything about your work on the music library in macOS this year? Thank you again for all your work on MusicKit! I’m planning to get the last of the load-bearing MediaPlayer code out of my codebase later this year. Hooray!

Is there anyway that we as users can set apps to not be able to play audio? There have been times that I want to listen to music and then scroll specific social media apps, and the social media apps keep interrupt the audio. I made a feedback related to this last year: FB18494861

My macOS app provides real-time microphone level monitoring, and I also have a macOS widget that shows recent audio level status. Since WidgetKit widgets are snapshot/timeline based rather than continuously live SwiftUI views, is there any new capability in macOS 27 that allows a widget to show more responsive or animated audio-level changes? If true real-time animation is still not supported in widgets, what is Apple’s recommended approach for displaying frequently changing signal levels, such as audio meters, in a macOS widget?

I’ve noticed that CATap appears to scale the stereo/mono mixdown based on the number of channels in the output audio device. For example, if the selected output device has 4 channels, the tapped audio is reduced by half. This behavior does not seem to be documented anywhere, but I’m assuming it may have been done to reduce the chance of clipping when downmixing. The issue is that users are complaining that the audio level is too low. Is there any workaround for this that does not involve tapping every audio device separately and doing my own mixdown? Ideally, I’d like to keep using the system tap/mix behavior but avoid the automatic level reduction, or at least compensate for it in a reliable way.

My app allows users to transcribe and translate audio from other apps running on your Mac. I currently use ScreenCaptureKit to do this, and just ignore the video stream and only use the audio stream. This works, but it's awkward because it requires I ask users for screen recording permission, even though all I want is the audio, not the screen. Is there a better way to do this in modern macOS releases? Thanks!

In the keynote (or maybe SOTU) they demo'd a new text dictation system. I'm wondering if this is being added to the Speech framework as a new transcription mechanism available to 3rd party apps. I couldn't find any videos referring to updates to the Speech framework.

I added logging in an AudioServerPlugin to trace the arrival and removal of a particular client in the AddDeviceClient() and RemoveDeviceClient() methods. To my surprise the app was removed and added as a client often, even when it was in a steady state, for example when using the Audio MIDI Setup to change sample rates. Is this expected behavior, and if so, are there any explanations about this.

In the keynote (or maybe it was SOTU) they demo'd a new more expressive Siri voice. Is that voice going to be available to 3rd party apps to use for text-to-speech in our apps?

I have an app where multiple audio files of the same length are played and they need to be in sync to produce the full song. My current setup syncs each track to the same position every time the user pauses or plays the song. However it is super laggy when the Audio is played either for the first time or when unpausing/pausing. What's the best way to do this?

I make a Mac app where users can transcribe and translate audio from other apps. I want to bring this to iOS, and I'm wondering what APIs are available (if any) to do this. On MacOS I use ScreenCaptureKit currently, but that's not available on iOS.

Hi, I encounter a dead-lock with Apple AUGraphicEQ. I have attached backtrace: thread #44, name = 'com.apple.audio.toolbox.AUScheduledParameterRefresher' frame #0: 0x0000000182059bb0 libsystem_kernel.dylib`semaphore_wait_trap + 8 frame #1: 0x000000018e9e7e00 caulk`caulk::semaphore::timed_wait(double) + 224 frame #2: 0x000000018e9e7cac caulk`caulk::concurrent::details::worker_thread::run() + 32 frame #3: 0x000000018e9e794c caulk`void* caulk::thread_proxy<std::__1::tuple<caulk::thread::attributes, void (caulk::concurrent::details::worker_thread::*)(), std::__1::tuple<caulk::concurrent::details::worker_thread*>>>(void*) + 96 frame #4: 0x000000018209dc58 libsystem_pthread.dylib`_pthread_start + 136 regards, Joël

I'm trying to create an app that can access a users Apple Music library to play music (the app will do far more than that when finished) but I am unable to generate a key. When I go to Register a New Key, it says: "There are no identifiers available that can be associated with the key" under Media Services (MusicKit, ShazamKit, Apple Music Feed). MusicKit is already checked under App Services in Certificates, Identifiers & Profiles. I've been stuck on this for two days. Please help!

I am experimenting with SpeechTranscriber and am curious if I can get quicker results when using buffered audio, rather than a file. The use case is a voice ordering experience for a restaurant. When I've been playing with it, it takes about 3 seconds for faster results and 7-8 seconds for accurate results. Is there any way to bring this down a bit? In this WWDC demo, the results appear nearly instantaneously. I'm curious how to replicate this in my app. I presume DicationTranscriber is faster, but how is siri detecting when the user stops speaking? Is it custom code, or is it using SpeechDetector? I tried using SpeechDetector with SpeechTranscriber but the detector didn't emit any results and seemed to slow down the results of SpeechTranscriber. I also assumed SpeechTranscriber makes more sense than DictationTranscriber in this use case, but want to confirm.

Have there been any changes in macOS 27 or iOS 27 that improve real-time audio level monitoring, WidgetKit updates, Live Activities, or audio route change handling for professional monitoring applications?

Why voice processing enabled on AVAudioInputNode makes output audio noticable lower than without it and how to overcome it using voice processing enabled

In iOS 27, when a user has configured a custom AirPods EQ (three-band lows/mids/highs), does that EQ apply on top of audio output from AVAudioEngine in my app, or only to media playback through AVPlayer/MPMusicPlayerController? My app generates therapeutic frequencies (e.g. 528 Hz pure tones) where precise frequency output matters - is there any API to detect or opt out of the system AirPods EQ?

For a sleep app running 8–12 hours in background, is AVAudioSourceNode with a real-time render block more power-efficient than looping a pre-encoded audio file via AVAudioPlayerNode? I want to migrate from files to procedural synthesis but not at the cost of battery. What does Instruments / Energy Log show as the typical CPU overhead difference, and is there Apple guidance on this trade-off?