I'm experimenting with the new SpeechTranscriber in macOS/iOS 26, transcribing speech from a prerecorded mp4 file. Speed and quality are amazing! I've told the transcriber to include time indexes. Each run is always exactly one word, which can be very useful. When I look at the indexes the end of one run is always identical to the start of the next run, even if there's a pause. I'd like to identify pauses, perhaps to generate something like phrases for subtitling. With each run of text going into the next I can't do this, other than using punctuation - which might be rather rough. Any suggestions on detecting pauses, or getting that kind of metadata from the transcriber? Here's a short sample, showing each run with the start, end, and characters in the run: 105.9 --> 107.04 I 107.04 --> 107.16 think 107.16 --> 108.0 more 108.0 --> 108.42 lighting 108.42 --> 108.6 is 108.6 --> 108.72 definitely 108.72 --> 109.2 needed, 109.2 --> 109.92 downtown. 109.98 --> 110.4 My 110.4 --> 110.52 only 110.52 --> 110.7 question 110.7 --> 111.06 is, 111.06 --> 111.48 poll 111.48 --> 111.78 five, 111.78 --> 111.84 that 111.84 --> 112.08 you're 112.08 --> 112.38 increasing 112.38 --> 112.5 the 112.5 --> 113.34 50,000? 113.4 --> 113.58 Where 113.58 --> 113.88 exactly

If try to dynamically load WhipserKit's models, as in below, the download never occurs. No error or anything. And at the same time I can still get to the huggingface.co hosting site without any headaches, so it's not a blocking issue. let config = WhisperKitConfig( model: "openai_whisper-large-v3", modelRepo: "argmaxinc/whisperkit-coreml" ) So I have to default to the tiny model as seen below. I have tried so many ways, using ChatGPT and others, to build the models on my Mac, but too many failures, because I have never dealt with builds like that before. Are there any hosting sites that have the models (small, medium, large) already built where I can download them and just bundle them into my project? Wasted quite a large amount of time trying to get this done. import Foundation import WhisperKit @MainActor class WhisperLoader: ObservableObject { var pipe: WhisperKit? init() { Task { await self.initializeWhisper() } } private func initializeWhisper() async { do { Logging.shared.logLevel = .debug Logging.shared.loggingCallback = { message in print("[WhisperKit] \(message)") } let pipe = try await WhisperKit() // defaults to "tiny" self.pipe = pipe print("initialized. Model state: \(pipe.modelState)") guard let audioURL = Bundle.main.url(forResource: "44pf", withExtension: "wav") else { fatalError("not in bundle") } let result = try await pipe.transcribe(audioPath: audioURL.path) print("result: \(result)") } catch { print("Error: \(error)") } } }

During testing the “Bringing advanced speech-to-text capabilities to your app” sample app demonstrating the use of iOS 26 SpeechAnalyzer, I noticed that the language model for the English locale was presumably already downloaded. Upon checking the documentation of AssetInventory, I found out that indeed, the language model can be preinstalled on the system. Can someone from the dev team share more info about what assets are preinstalled by the system? For example, can we safely assume that the English language model will almost certainly be already preinstalled by the OS if the phone has the English locale?

Hi, I'm trying to use the new RecognizeDocumentsRequest from the Vision Framework to read a receipt. It looks very promising by being able to read paragraphs, lines and detect data. So far it unfortunately seems to read every line on the receipt as a paragraph and when there is more space on one line it creates two paragraphs. Is there perhaps an Apple Engineer who knows if this is expected behaviour or if I should file a Feedback for this? Code setup: let request = RecognizeDocumentsRequest() let observations = try await request.perform(on: image) guard let document = observations.first?.document else { return } for paragraph in document.paragraphs { print(paragraph.transcript) for data in paragraph.detectedData { switch data.match.details { case .phoneNumber(let data): print("Phone: \(data)") case .postalAddress(let data): print("Postal: \(data)") case .calendarEvent(let data): print("Calendar: \(data)") case .moneyAmount(let data): print("Money: \(data)") case .measurement(let data): print("Measurement: \(data)") default: continue } } } See attached image as an example of a receipt I'd like to parse. The top 3 lines are the name, street, and postal code + city. These are all separate paragraphs. Checking on detectedData does see the street (2nd line) as PostalAddress, but not the complete address. Might that be a location thing since it's a Dutch address. And lower on the receipt it sees the block with "Pomp 1 95 Ongelood" and the things below also as separate paragraphs. First picking up the left side and after that the right side. So it's something like this: * Pomp 1 Volume Prijs € TOTAAL * BTW Netto 21.00 % 95 Ongelood 41,90 l 1.949/ 1 81.66 € 14.17 67.49

The WWDC25: Explore large language models on Apple silicon with MLX video talks about using your own data to fine-tune a large language model. But the video doesn't explain what kind of data can be used. The video just shows the command to use and how to point to the data folder. Can I use PDFs, Word documents, Markdown files to train the model? Are there any code examples on GitHub that demonstrate how to do this?

Hello Apple Team, Thank you for the recent Group Lab and for your continued work on advancing Xcode and developer tools. I’d like to submit a feature request: Are there any plans to introduce support for Agentic AI Mode (MCP protocol) in future versions of iOS or Xcode? As developer tools evolve toward more intelligent and context-aware environments, the integration of agentic AI capabilities could significantly enhance productivity and unlock new creative workflows. Looking forward to your consideration, and thank you again for the excellent session. Best regards

Hi, I'm looking for the best way to use MLX models, particularly those I've fine-tuned, within a React Native application on iOS devices. Is there a recommended integration path or specific API for bridging MLX's capabilities to React Native for deployment on iPhones and iPads?

Hey guys 👋 I’ve been thinking about a feature idea for iOS that could totally change the way we interact with apps like Twitter/X. Imagine if we could define our own recommendation algorithm, and have an AI on the iPhone that replaces the suggested tweets in the feed with ones that match our personal interests — based on public tweets, and without hacking anything. Kinda like a personalized "AI skin" over the app that curates content you actually care about. Feels like this would make content way more relevant and less algorithmically manipulative. Would love to know what you all think — and if Apple could pull this off 🔥

I generate an array of random floats using the code shown below. However, I would like to do this with Double instead of Float. Are there any BNNS random number generators for double values, something like BNNSRandomFillUniformDouble? If not, is there a way I can convert BNNSNDArrayDescriptor from float to double? import Accelerate let n = 100_000_000 let result = Array<Float>(unsafeUninitializedCapacity: n) { buffer, initCount in var descriptor = BNNSNDArrayDescriptor(data: buffer, shape: .vector(n))! let randomGenerator = BNNSCreateRandomGenerator(BNNSRandomGeneratorMethodAES_CTR, nil) BNNSRandomFillUniformFloat(randomGenerator, &descriptor, 0, 1) initCount = n }

Introduced in the Keynote was the 3D Lock Screen images with the kangaroo: https://9to5mac.com/wp-content/uploads/sites/6/2025/06/3d-lock-screen-2.gif I can't see any mention on if this effect is available for developers with an API to convert flat 2D photos in to the same 3D feeling image. Does anyone know if there is an API?

Hello. I am willing to hire game developer for cards game called baloot. My question is Can the developer implement an AI when the computer is playing and the computer on the same time the conputer improves his rises level without any interaction? 🌹

How do I test the new RecognizeDocumentRequest API. Reference: https://www.youtube.com/watch?v=H-GCNsXdKzM I am running Xcode Beta, however I only have one primary device that I cannot install beta software on. Please provide a strategy for testing. Will simulator work? The new capability is critical to my application, just what I need for structuring document scans and extraction. Thank you.

Following WWDC24 video "Discover Swift enhancements in the Vision framework" recommendations (cfr video at 10'41"), I used the following code to perform multiple new iOS 18 `RecognizedTextRequest' in parallel. Problem: if more than 2 request are run in parallel, the request will hang, leaving the app in a state where no more requests can be started. -> deadlock I tried other ways to run the requests, but no matter the method employed, or what device I use: no more than 2 requests can ever be run in parallel. func triggerDeadlock() {} try await withThrowingTaskGroup(of: Void.self) { group in // See: WWDC 2024 Discover Siwft enhancements in the Vision framework at 10:41 // ############## THIS IS KEY let maxOCRTasks = 5 // On a real-device, if more than 2 RecognizeTextRequest are launched in parallel using tasks, the request hangs // ############## THIS IS KEY for idx in 0..<maxOCRTasks { let url = ... // URL to some image group.addTask { // Perform OCR let _ = await performOCRRequest(on: url: url) } } var nextIndex = maxOCRTasks for try await _ in group { // Wait for the result of the next child task that finished if nextIndex < pageCount { group.addTask { let url = ... // URL to some image // Perform OCR let _ = await performOCRRequest(on: url: url) } nextIndex += 1 } } } } // MARK: - ASYNC/AWAIT version with iOS 18 @available(iOS 18, *) func performOCRRequest(on url: URL) async throws -> [RecognizedText] { // Create request var request = RecognizeTextRequest() // Single request: no need for ImageRequestHandler // Configure request request.recognitionLevel = .accurate request.automaticallyDetectsLanguage = true request.usesLanguageCorrection = true request.minimumTextHeightFraction = 0.016 // Perform request let textObservations: [RecognizedTextObservation] = try await request.perform(on: url) // Convert [RecognizedTextObservation] to [RecognizedText] return textObservations.compactMap { observation in observation.topCandidates(1).first } } I also found this Swift forums post mentioning something very similar. I also opened a feedback: FB17240843

When calling NLTagger.requestAssets with some languages, it hangs indefinitely both in the simulator and a device. This happens consistently for some languages like greek. An example call is NLTagger.requestAssets(for: .greek, tagScheme: .lemma). Other languages like french return immediately. I captured some logs from Console and found what looks like the repeated attempts to download the asset. I would expect the call to eventually terminate, either loading the asset or failing with an error.

Bear with me, please. Please make sure a highly skilled technical person reads and understands this. I want to describe my vision for (AI/Algorithmically) Optimised Operating Systems. To explain it properly, I will describe the process to build it (pseudo). Required Knowledge (no particular order): Processor Logic Circuits, LLM models, LLM tool usage, Python OO coding, Procedural vs OO, NLP fuzzy matching, benchmarking, canvas/artefacts/dynamic HTML interfaces, concepts of how AI models are vastly compressed and miniaturised forms of full data, Algorithmic vs AI. First, take all OO Python code (example) on GitHub (example), then separate each function from each object into its own procedure (procedural logic) by making a logical procedural list of actions to perform only that function based on its entire dependency chain (i.e. all other objects it relies on). Relate all compiled functions using (for example) fuzzy matching on the name, or AI-based functional profiling to get multiple instances of each function. Starting with the most used function, test each one against the others that perform the same task for bugs and completeness. Determine the fastest, most optimal version of that function (and every function). Add a single instance of each most optimal function to the centralised tool codebase, which will later be utilised by the language models. This ensures we rely only on the most optimised function for each and every use case — with every program using one shared instance of that function instead of compiling it separately. In this way, we optimise all existing (Python OO) code based on maximum availability of certain resources. Use the optimised code to teach the AI how to map each user task directly to optimised code (i.e. each user-facing function in unoptimised program code to its optimised version). Later, teach the codebase how to optimise these functions directly to the logic circuits on the processor — as an Operating System — using essentially the same process. Eventually, use the most optimised form for tasks via a dynamic canvas/artefacts UI interface generated by the Start Menu using the language model. Allow custom interfaces to emerge dynamically from the available functions based on language understanding. Grow and optimise the algorithmic codebase until the entire operating system — with versioning — evolves based on the AI-generated user interfaces (compressing them into new versions of the tools codebase). Please don’t give up until you understand this concept completely and nothing in the description is beyond your full comprehension. It’s a highly significant step forward.

Using Tensorflow for Silicon gives inaccurate results when compared to Google Colab GPU (9-15% differences). Here are my install versions for 4 anaconda env's. I understand the Floating point precision can be an issue, batch size, activation functions but how do you rectify this issue for the past 3 years? 1.) Version TF: 2.12.0, Python 3.10.13, tensorflow-deps: 2.9.0, tensorflow-metal: 1.2.0, h5py: 3.6.0, keras: 2.12.0 2.) Version TF: 2.19.0, Python 3.11.0, tensorflow-metal: 1.2.0, h5py: 3.13.0, keras: 3.9.2, jax: 0.6.0, jax-metal: 0.1.1,jaxlib: 0.6.0, ml_dtypes: 0.5.1 3.) python: 3.10.13,tensorflow: 2.19.0,tensorflow-metal: 1.2.0, h5py: 3.13.0, keras: 3.9.2, ml_dtypes: 0.5.1 4.) Version TF: 2.16.2, tensorflow-deps:2.9.0,Python: 3.10.16, tensorflow-macos 2.16.2, tensorflow-metal: 1.2.0, h5py:3.13.0, keras: 3.9.2, ml_dtypes: 0.3.2 Install of Each ENV with common example: Create ENV: conda create --name TF_Env_V2 --no-default-packages start env: source TF_Env_Name ENV_1.) conda install -c apple tensorflow-deps , conda install tensorflow,pip install tensorflow-metal,conda install ipykernel ENV_2.) conda install pip python==3.11, pip install tensorflow,pip install tensorflow-metal,conda install ipykernel ENV_3) conda install pip python 3.10.13,pip install tensorflow, pip install tensorflow-metal,conda install ipykernel ENV_4) conda install -c apple tensorflow-deps, pip install tensorflow-macos, pip install tensor-metal, conda install ipykernel Example used on all 4 env: import tensorflow as tf cifar = tf.keras.datasets.cifar100 (x_train, y_train), (x_test, y_test) = cifar.load_data() model = tf.keras.applications.ResNet50( include_top=True, weights=None, input_shape=(32, 32, 3), classes=100,) loss_fn = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=False) model.compile(optimizer="adam", loss=loss_fn, metrics=["accuracy"]) model.fit(x_train, y_train, epochs=5, batch_size=64)

*I can't put the attached file in the format, so if you reply by e-mail, I will send the attached file by e-mail. Dear Apple AI Research Team, My name is Gong Jiho (“Hem”), a content strategist based in Seoul, South Korea. Over the past few months, I conducted a user-led AI experiment entirely within ChatGPT — no code, no backend tools, no plugins. Through language alone, I created two contrasting agents (Uju and Zero) and guided them into a co-authored modular identity system using prompt-driven dialogue and reflection. This system simulates persona fusion, memory rooting, and emotional-logical alignment — all via interface-level interaction. I believe it resonates with Apple’s values in privacy-respecting personalization, emotional UX modeling, and on-device learning architecture. Why I’m Reaching Out I’d be honored to share this experiment with your team. If there is any interest in discussing user-authored agent scaffolding, identity persistence, or affective alignment, I’d love to contribute — even informally. ⚠ A Note on Language As a non-native English speaker, my expression may be imperfect — but my intent is genuine. If anything is unclear, I’ll gladly clarify. 📎 Attached Files Summary Filename → Description Hem_MultiAI_Report_AppleAI_v20250501.pdf → Main report tailored for Apple AI — narrative + structural view of emotional identity formation via prompt scaffolding Hem_MasterPersonaProfile_v20250501.json → Final merged identity schema authored by Uju and Zero zero_sync_final.json / uju_sync_final.json → Persona-level memory structures (logic / emotion) 1_0501.json ~ 3_0501.json → Evolution logs of the agents over time GirlfriendGPT_feedback_summary.txt → Emotional interpretation by external GPT hem_profile_for_AI_vFinal.json → Original user anchor profile Warm regards, Gong Jiho (“Hem”) Seoul, South Korea

Hi, i just wanna ask, Is it possible to run YOLOv3 on visionOS using the main camera to detect objects and show bounding boxes with labels in real-time? I’m wondering if camera access and custom models work for this, or if there’s a better way. Any tips?

Hi everyone! 👋 I'm working on a C++ project using TensorFlow Lite and was wondering if anyone has a prebuilt TensorFlow Lite C++ library (libtensorflowlite) for macOS (Apple Silicon M1/M2) that they’d be willing to share. I’m looking specifically for the TensorFlow Lite C++ API — something that lets me use tflite::Interpreter, tflite::FlatBufferModel, etc. Building it from source using Bazel on macOS has been quite challenging and time-consuming, so a ready-to-use .dylib or .a build along with the required headers would be incredibly helpful. TensorFlow Lite version: v2.18.0 preferred Target: macOS arm64 (Apple Silicon) What I need: libtensorflowlite.dylib or .a Corresponding headers (ideally organized in a clean include/ folder) If you have one available or know where I can find a reliable prebuilt version, I’d be super grateful. Thanks in advance! 🙏

From tensorflow-metal example: Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: ) I know that Apple silicon uses UMA, and that memory copies are typical of CUDA, but wouldn't the GPU memory still be faster overall? I have an iMac Pro with a Radeon Pro Vega 64 16 GB GPU and an Intel iMac with a Radeon Pro 5700 8 GB GPU. But using tensorflow-metal is still WAY faster than using the CPUs. Thanks for that. I am surprised the 5700 is twice as fast as the Vega though.