I'm able to get text to speech to audio file using the following code for iOS 12 iPhone 8 to create a car file: audioFile = try AVAudioFile( forWriting: saveToURL, settings: pcmBuffer.format.settings, commonFormat: .pcmFormatInt16, interleaved: false) where pcmBuffer.format.settings is: [AVAudioFileTypeKey: kAudioFileMP3Type, AVSampleRateKey: 48000, AVEncoderBitRateKey: 128000, AVNumberOfChannelsKey: 2, AVFormatIDKey: kAudioFormatLinearPCM] However, this code does not work when I run the app in iOS 18 on iPhone 13 Pro Max. The audio file is created, but it doesn't sound right. It has a lot of static and it seems the speech is very low pitch. Can anyone give me a hint or an answer?

On Apple TV 4K 3rd generation, with tvOS 26 beta 2, when two HomePod 2 are paired to the device, music and movie sources with Dolby Atmos can only be listened to in stereo. dolby atmos not supported

I'm using an AUGraph to mix audio from different sources for a real time streaming application. Whenever the audio device used as the graph's output device is also the Mac's default output device, the measured latency increases by about 35 milliseconds for wired devices. Any idea why this is? Is there a way around this besides nagging the user to not the use the system output in our app?

Hi. I want to read ADPCM encoded audio data, coming from an external device to my Mac via serial port (/dev/cu.usbserial-0001) as 256 byte chunks, and feed it into an audio player. So far I am using Swift and SwiftSerial (GitHub - mredig/SwiftSerial: A Swift Linux and Mac library for reading and writing to serial ports. 3) to get the data via serialPort.asyncBytes() into a AsyncStream but I am struggling to understand how to feed the stream to a AVAudioPlayer or similar. I am new to Swift and macOS audio development so any help to get me on the right track is greatly appreciated. Thx

Environment→ ・Device: iPad 10th generation ・OS：**iOS18.3.2 I'm using AVAudioSession to record sound in my application. But I recently came to realize that when the app starts a recording session on a tablet, OS automatically sets the tablet volume to 50% and when after recording ends, it doesn't change back to the previous volume level before starting the recording. So I would like to know whether this is an OS default behavior or a bug? If it's a default behavior, I much appreciate if I can get a link to the documentation.

Hi all, I've developed an audio DSP application in C++ using AudioToolbox and CoreAudio on MacOS 14.4.1 with Xcode 15. I use an AudioQueue for input and another for output. This works great. I'm now adding real-time audio analysis eg spectral analysis. I want this to run independently of my audio processing so it can not interfere with audio playback. Taps on AudioQueues seem to be a good way of doing this... Since the analytics won't modify the audio data, I am using a Siphon Tap by setting the AudioQueueProcessingTapFlags to kAudioQueueProcessingTap_PreEffects | kAudioQueueProcessingTap_Siphon; This works fine on my output queue. However, on my input queue the Tap callback is called once and then a EXC_BAD_ACCESS occurs - screen shot below. NB: I believe that a callback should only call AudioQueueProcessingTapGetSourceAudio when not using a Siphon, so I don't call it. Relevant code: AudioQueueProcessingTapCallback tap_callback) { // Makes an audio tap for a queue void * tap_data_ptr = NULL; AudioQueueProcessingTapFlags tap_flags = kAudioQueueProcessingTap_PostEffects | kAudioQueueProcessingTap_Siphon; uint32_t max_frames = 0; AudioStreamBasicDescription asbd; AudioQueueProcessingTapRef tap_ref; OSStatus status = AudioQueueProcessingTapNew(queue_ref, tap_callback, tap_data_ptr, tap_flags, &max_frames, &asbd, &tap_ref); if (status != noErr) printf("Error while making Tap\n"); else printf("Successfully made tap\n"); } void tapper(void * tap_data, AudioQueueProcessingTapRef tap_ref, uint32_t number_of_frames_in, AudioTimeStamp * ts_ptr, AudioQueueProcessingTapFlags * tap_flags_ptr, uint32_t * number_of_frames_out_ptr, AudioBufferList * buf_list) { // Callback function for audio queue tap printf("Tap callback"); }``` Image of exception stack provided by Xcode: ![]("https://developer.apple.com/forums/content/attachment/27479e8d-a118-459b-aa2d-7e30528910e3" "title=Screenshot 2025-06-14 at 1.29.14 PM.png;width=932;height=562") What have I missed? Appreciate any help you learned folks may be able to provide. Best, Geoff.

Hello, I am a deaf-blind wheelchair user, and I program in Swift using a braille display. I’m reaching out for your help on an issue I’ve been struggling to solve. Basically, when I extract a CMSampleBuffer from an AVAsset of a video, it comes with the Audio Format ID as Linear PCM. However, when I try to pass this CMSampleBuffer to write another video using AVAssetWriter, the video ends up muted. The audio settings of the output video are configured to MPEG-4 AAC, but the input CMSampleBuffer has the Audio Format ID as Linear PCM. I would like to request an extension for CMSampleBuffer that converts Linear PCM audio to MPEG-4 AAC. I’ve searched extensively and couldn’t find anything. Looking forward to your help. Thank you.

When using the Apple Devices to sync Apple Music to iPhone where is the Apple Devices backup being written to? Apple Devices->music->sync. Not trying to backup the iPhone via Apple Devices app.

Hey, I am fairly new to working with AVFoundation etc. As far as I could research on my own, if I want to get metadata from let's say a .m4a audio file, I have to get the data and then create an AVAsset. My files are all on local servers and therefore I would not be able to just pass in the URL. The extraction of the metadata works fine - however those AVAssets create a huge overhead in storage consumption. To my knowledge the data instances of each audio file and AVAsset should only live inside the function I call to extract the metadata, however those data/AVAsset instances still live on on storage as I can clearly see that the app's file size increases by multiple Gigabytes (equal to the library size I test with). However, the only data that I purposefully save with SwiftData is the album artwork. Is this normal behavior for AVAssets or am I missing some detail? PS. If I forgot to mention something important, please ask. This is my first ever post, so I'm not too sure what is worth mentioning. Thank you in advance! Denis

Hello there! Is there any list of voices that are always available on iOS/iPadOS devices? It seems that AVSpeechSynthesisVoice(identifier: "com.apple.voice.compact.en-US.Samantha") is always available on all devices. I thought that AVSpeechSynthesisVoice(identifier: "com.apple.ttsbundle.siri_Nicky_en-US_compact") and AVSpeechSynthesisVoice(identifier: "com.apple.ttsbundle.siri_Aaron_en-US_compact") were available by default on certain newer devices. Is this true? I also noticed that on the same iPad where I was using those 2 voices (Nicky and Aaron) - when I updated to the iPadOS 26 beta, those voices were no longer available. Any information you can share about which voices should be reliably available on which devices would be extremely helpful for our development. Thanks so much!

Getting MatchError "MATCH_ATTEMPT_FAILED" everytime when matchstream on Android Studio Java+Kotlin project. My project reads the samples from the mic input using audioRecord class and sents them to the Shazamkit to matchstream. I created a kotlin class to handle to Shazamkit. The audioRecord is build to be mono and 16 bit. My Kotlin Class class ShazamKitHelper { val shazamScope = CoroutineScope(Dispatchers.IO + SupervisorJob()) lateinit var streaming_session: StreamingSession lateinit var signature: Signature lateinit var catalog: ShazamCatalog fun createStreamingSessionAsync(developerTokenProvider: DeveloperTokenProvider, readBufferSize: Int, sampleRate: AudioSampleRateInHz ): CompletableFuture<Unit>{ return CompletableFuture.supplyAsync { runBlocking { runCatching { shazamScope.launch { createStreamingSession(developerTokenProvider,readBufferSize,sampleRate) }.join() }.onFailure { throwable -> }.getOrThrow() } } } private suspend fun createStreamingSession(developerTokenProvider:DeveloperTokenProvider,readBufferSize: Int,sampleRateInHz: AudioSampleRateInHz) { catalog = ShazamKit.createShazamCatalog(developerTokenProvider) streaming_session = (ShazamKit.createStreamingSession( catalog, sampleRateInHz, readBufferSize ) as ShazamKitResult.Success).data } fun startMatching() { val audioData = sharedAudioData ?: return // Return if sharedAudioData is null CoroutineScope(Dispatchers.IO).launch { runCatching { streaming_session.matchStream(audioData.data, audioData.meaningfulLengthInBytes, audioData.timestampInMs) }.onFailure { throwable -> Log.e("ShazamKitHelper", "Error during matchStream", throwable) } } } @JvmField var sharedAudioData: AudioData? = null; data class AudioData(val data: ByteArray, val meaningfulLengthInBytes: Int, val timestampInMs: Long) fun startListeningForMatches() { CoroutineScope(Dispatchers.IO).launch { streaming_session.recognitionResults().collect { matchResult -> when (matchResult) { is MatchResult.Match -> { val match = matchResult.matchedMediaItems println("Match found: ${match.get(0).title} by ${match.get(0).artist}") } is MatchResult.NoMatch -> { println("No match found") } is MatchResult.Error -> { val error = matchResult.exception println("Match error: ${error.message}") } } } } } } My code in java reads the samples from a thread: shazam_create_session(); while (audioRecord.getRecordingState() == AudioRecord.RECORDSTATE_RECORDING){ if (shazam_session_created){ byte[] buffer = new byte[288000];//max_shazam_seconds * sampleRate * 2]; audioRecord.read(buffer,0,buffer.length,AudioRecord.READ_BLOCKING); helper.sharedAudioData = new ShazamKitHelper.AudioData(buffer,buffer.length,System.currentTimeMillis()); helper.startMatching(); if (!listener_called){ listener_called = true; helper.startListeningForMatches(); } } else{ SystemClock.sleep(100); } } private void shazam_create_session() { MyDeveloperTokenProvider provider = new MyDeveloperTokenProvider(); AudioSampleRateInHz sample_rate = AudioSampleRateInHz.SAMPLE_RATE_48000; if (sampleRate == 44100) sample_rate = AudioSampleRateInHz.SAMPLE_RATE_44100; CompletableFuture<Unit> future = helper.createStreamingSessionAsync(provider, 288000, sample_rate); future.thenAccept(result -> { shazam_session_created = true; }); future.exceptionally(throwable -> { Toast.makeText(mine, "Failure", Toast.LENGTH_SHORT).show(); return null; }); } I Implemented the developer token in java as follows public static class MyDeveloperTokenProvider implements DeveloperTokenProvider { DeveloperToken the_token = null; @NonNull @Override public DeveloperToken provideDeveloperToken() { if (the_token == null){ try { the_token = generateDeveloperToken(); return the_token; } catch (NoSuchAlgorithmException | InvalidKeySpecException e) { throw new RuntimeException(e); } } else{ return the_token; } } @NonNull private DeveloperToken generateDeveloperToken() throws NoSuchAlgorithmException, InvalidKeySpecException { PKCS8EncodedKeySpec priPKCS8 = new PKCS8EncodedKeySpec(Decoders.BASE64.decode(p8)); PrivateKey appleKey = KeyFactory.getInstance("EC").generatePrivate(priPKCS8); Instant now = Instant.now(); Instant expiration = now.plus(Duration.ofDays(90)); String jwt = Jwts.builder() .header().add("alg", "ES256").add("kid", keyId).and() .issuer(teamId) .issuedAt(Date.from(now)) .expiration(Date.from(expiration)) .signWith(appleKey) // Specify algorithm explicitly .compact(); return new DeveloperToken(jwt); } }

Your draft looks great! Here's a refined version with the iOS 17 comparison emphasized and slightly better flow: Hi Apple Engineers and fellow developers, I'm experiencing a critical regression with ShazamKit's background operation on iOS 18. ShazamKit's SHManagedSession stops identifying songs in the background after approximately 20 seconds on iOS 18, while the exact same code works perfectly on iOS 17. The behavior is consistent: the app works perfectly in the foreground, but when backgrounded or device is locked, it initially works for about 20 seconds then stops identifying new songs. The microphone indicator remains active suggesting audio access is maintained, but ShazamKit doesn't send identified songs in the background until you open the app again. Detection immediately resumes when bringing the app to foreground. My technical setup uses SHManagedSession for continuous matching with background modes properly configured in Info.plist including audio mode, and Background App Refresh enabled. I've tested this on physical devices running iOS 18.0 through 18.5 with the same results across all versions. The exact same code running on iOS 17 devices works flawlessly in the background. To reproduce: initialize SHManagedSession and start matching, begin song identification in foreground, background the app or lock device, play different songs which are initially detected for about 20 seconds, then after the timeout period new songs are no longer identified until you bring the app to foreground. This regression has impacted my production app as users who rely on continuous background music identification are experiencing a broken feature. I submitted this as Feedback ID FB15255903 last September with no solution so far. I've created a minimal demo project that reproduces this issue: https://github.com/tfmart/ShazamKitBackground Has anyone else experienced this ShazamKit background regression on iOS 18? Are there any known workarounds or alternative approaches? Given the time this issue has persisted, could we please get acknowledgment of this regression, expected timeline for a fix, or any recommended workarounds? Testing environment is Xcode 16.0+ on iOS 18.0-18.5 across multiple physical device models. Any guidance would be greatly appreciated.

Hi everyone, I wanted to bring up a question about Core Audio and its potential for future updates or improvements, specifically regarding latency optimization. As someone who relies on Core Audio for real-time audio processing, any enhancements in this area would be incredibly beneficial for professionals in the industry. Does anyone know if Apple has shared any plans or updates regarding Core Audio’s performance, particularly for low-latency applications? I’d appreciate any insights or advice from the community! Thanks so much! Best, Michael

I have a music app I'm developing and having a weird issue where I can see now playing info for every other platform than tvOS. As far as I can tell I have correctly configured the MPNowPlayingInfoCenter MPNowPlayingInfoCenter.default().nowPlayingInfo = nowPlayingInfo MPNowPlayingInfoCenter.default().playbackState = .playing Are there any extra requirements to get my app's now-playing info showing in control center on tvOS? Another strange issue that might be related is I can use the apple TV remote to pause audio but not resume playback, so I feel like there's something I'm missing about registering audio playback on tvOS specifically.

I've got a web app built with MusicKit that displays a list of songs. I have player controls for play, pause, skip next, skip, previous, toggle shuffle and set repeat mode. All of these work by using music. The play button, when nothing is playing and nothing is in the queue, will enqueue all the tracks and start playing with the below, for example: await music.setQueue({ songs, startPlaying: true }); I've implemented a progress slider based on feedback from the "playbackProgressDidChange" listener. Now, how in the world can I set the volume? This seems like it should be simple, but I am at a complete loss here. The docs say: "The volume of audio playback, which is set directly on the HTMLMediaElement as the HTMLMediaElement.volume property. This value ranges between 0, which would be muting the audio, and 1, which would be the loudest possible." Given that all my controls work off the music instance, I don't understand how I can do that. In this video from WWDC 2022, music web components are touched on briefly. These are also documented very sparsely. The volume docs are here. For the life of me, I can't even get the volume web component to display in the UI. It appears that MusicKit Web is hobbled compared to the native implementation, but surely adjusting volume shouldn't be that hard right? I'd appreciate any insight on how to do this, including how to get web components to work (in a Next JS app). Thanks.

I'm building a streaming app on visionOS that can play sound from audio buffers each frame. The audio format has a bitrate of 48000, and each buffer has 480 samples. I noticed when calling audioPlayerNode.scheduleBuffer(audioBuffer) The memory keeps increasing at the speed of 0.1MB per second And at around 4 minutes, the node seems to be full of buffers and had a hard reset, at which point, the audio is stopped temporary with a memory change. see attached screenshot. However, if I call audioPlayerNode.scheduleBuffer(audioBuffer, at: nil, options: .interrupts) The memory leak issue is gone, but the audio is broken (sounds like been shortened). Below is the full code snippet, anyone knows how to fix it? @Observable final class MyAudioPlayer { private var audioEngine: AVAudioEngine = .init() private var audioPlayerNode: AVAudioPlayerNode = .init() private var audioFormat: AVAudioFormat? init() { audioEngine.attach(audioPlayerNode) audioEngine.connect(audioPlayerNode, to: audioEngine.mainMixerNode, format: nil) try? AVAudioSession.sharedInstance().setCategory(.playback, mode: .default) try? AVAudioSession.sharedInstance().setActive(true) audioEngine.prepare() try? audioEngine.start() audioPlayerNode.play() } // more code... /// callback every frame private func audioFrameCallback_Non_Interleaved(buf: UnsafeMutablePointer<Float>?, samples: Int) { guard let buf, let format = AVAudioFormat(commonFormat: .pcmFormatFloat32, sampleRate: 48000, channels: 2, interleaved: false), let audioBuffer = AVAudioPCMBuffer(pcmFormat: format, frameCapacity: AVAudioFrameCount(samples)) else { return } audioBuffer.frameLength = AVAudioFrameCount(samples) if let data = audioBuffer.floatChannelData { for channel in 0 ..< Int(format.channelCount) { for frame in 0 ..< Int(audioBuffer.frameLength) { data[channel][frame] = buf[frame * Int(format.channelCount) + channel] } } } // memory leak here audioPlayerNode.scheduleBuffer(audioBuffer) } }

When we tested the audio quality of our VoIP App, we found that when the iOS18.0 device was played with AirPods Pro 2, we could hear noises similar to peak clipping and distortion, especially when the sound source played was loud and high-pitched. Here is the device information we tested: Model: iPhone 16 Pro Max, iPhone 15 Pro System version: iOS 18.0 (22A3354) Bluetooth headset model: AirPods Pro 2 Bluetooth firmware version: 6F8 We tested multiple apps (including phone calls, FaceTime, Zoom, WeChat, Tencent Meeting), and they all had the above noise problem. We also found two phenomena: If we use the same iOS 18 device to connect HUAWEI FreeBuds Pro or FreeBuds 2, there is no such noise problem; If we use an iOS 17 device to connect to the same AirPods Pro 2 for testing, there is no such noise problem; Therefore, we suspect that it is caused by the compatibility problem between iOS 18.0 and AirPods firmware 6F8. The firmware version of our AirPods Pro 2 is 6F8, which was released on June 26, and iOS 18.0 was released on September 16. Maybe they are not very compatible. I hope that subsequent firmware updates can fix this problem.

Hi! I have a music app using AVAudioEngine. Right now, I have set it up to play multi channel tracks and show "Multichannel" in the volume controls. However, I am unable to figure out how to get it to use Dolby Atmos. Is there something that needs to be enabled? Is it even possible for AVAudioEngine? I saw some apps that are able of playing with Dolby Atmos, but they do not have EQ feature, so I'm guessing that they are not using AVAudioEngine.

I’ve encountered an issue when trying to transcribe audio during a SharePlay session in VisionOS. Specifically, the AVAudioSession appears to fail when sharing audio, preventing successful transcription. The problem seems related to AVAudioSession.sharedInstance() and using the .mixWithOthers option, which is supposed to enable multiple audio sources to coexist without interference. Here’s the relevant code snippet that throws the error: private static func prepareEngine() throws -> (AVAudioEngine, SFSpeechAudioBufferRecognitionRequest) { let audioEngine = AVAudioEngine() let request = SFSpeechAudioBufferRecognitionRequest() request.shouldReportPartialResults = true let audioSession = AVAudioSession.sharedInstance() try audioSession.setCategory(.playAndRecord, mode: .default, options: [.mixWithOthers, .allowBluetooth]) try audioSession.setActive(true, options: .notifyOthersOnDeactivation) let inputNode = audioEngine.inputNode let recordingFormat = inputNode.outputFormat(forBus: 0) inputNode.installTap(onBus: 0, bufferSize: 1024, format: recordingFormat) { (buffer: AVAudioPCMBuffer, when: AVAudioTime) in request.append(buffer) } audioEngine.prepare() try audioEngine.start() return (audioEngine, request) } The setup is designed to initialize an AVAudioEngine and a SFSpeechAudioBufferRecognitionRequest for real-time transcription, but fails within the SharePlay context. Notably, while .mixWithOthers is intended to handle concurrent audio sessions, it doesn’t appear to work as expected during SharePlay. The audioSession.setActive(true) line is where the setup typically fails, with no clear solution to proceed. Has anyone else faced similar issues with AVAudioSession and SharePlay in VisionOS? Any insights on how to manage audio sharing or transcription during a SharePlay session would be greatly appreciated! The specific error is: The operation couldn't be completed. (com.apple.coreaudio.avfaudio error 561145187.)

I have spent a long time refactoring lots of older Swift code to compile without error in Swift 6. The app is a v3 audio unit host and audio unit. Having installed Sonoma and XCode 16 I compile the code using Swift 6 and it compiles and runs without any warnings or errors. My host will load my AU no problem. LOGIC PRO is still the ONLY audio unit host that will load native Mac V3 audio units and so I like to test my code using Logic. In Sonoma with XCode 16... My AU passes the most stringent AUVAL tests both in terminal and Logic pro. If I compile the AU source in Swift 5 Logic will see the AU, load it and run it without problems. But when I compile the AU in Swift 6 Logic sees the AU, will scan it and verify it passes the tests but will not load the AU. In XCode I see a log message that a "helper application failed to run" but the debugger never connects to the AU and I don't think Logic even gets as far as instantiating the AU. So... what is causing this? I'm stumped.. Developing AUv3 is a brain-aching maze of undocumented hurdles and I'm hoping someone might have found a solution for this one. Meanwhile I guess my only option is to continue using the Swift 5 compiler. (appending a little note just to mention that all the DSP code is written in C/C++, Swift is used mainly for the user interface and also does some offline thready work )