I have an iOS app with multiple subclasses of UIViewControllers. There are many type of UIAlertControllers I might need to use based on user interaction, internet connection, and catching any other fatal errors. So I wrote the extension for UIViewController below, which works just fine. And I can call from any of my UIViewControllers as simply as: myErrors(error: MyErrors.e1.rawValue, title: "Internet Error", msg: "Unable to connect to Internet\nTry Again?") While this works, I do not know if it's proper to add an extension to UIViewController. Is this considered bad practice? Is there another way I should be pursuing this? extension UIViewController { func myErrors(error: MyErrors, title: String, msg: String) { var title = "" var message = "" switch error { case .e1: title = String(format: "%@", title) message = String(format: "Database Error %03d%@\n", error.rawValue, msg) case .e2: title = String(format: "%@", title) message = String(format: "Internet Error %03d%@\n", error.rawValue, msg) case .e3: title = String(format: "%@", title) message = String(format: "User Error %03d%@\n", error.rawValue, msg) } let alert = UIAlertController(title: title, message: message, preferredStyle: UIAlertController.Style.alert) switch error { case .e1: alert.addAction(UIAlertAction(title: "No", style: .init(rawValue: 0)!, handler: { (action: UIAlertAction!) in // ..log error //...proceed to code based on No .... })) alert.addAction(UIAlertAction(title: "Yes", style: .init(rawValue: 0)!, handler: { (action: UIAlertAction!) in // ..log error //...code based on Yes .... })) case .e2: // No user option availabe in this alert, just OK // ... do all logging of errors // proceed case .e3: // Add specific acctions to this error // ... do all logging of errors // proceed } self.present(alert, animated: true, completion: nil) } }

I created the playground below to answer my question "If I created a class instance using DispatchQueue.global().async would that class remain in its own asynchronous queue? Even if the main app called one of that classes methods, would that method run asynchronously compared to the main app? With the sleep line I discovered that the answer is "no." But I am curious if there is a legit way to do this? Or even if there is, it is considered bad programming? import UIKit class MyFunx : NSObject { var opsCount = 0 override init() { super.init() } func addValues (a: Int, b: Int) { let c = a + b opsCount += 1 sleep(1) } } var firstVar = 0 var secondVar = 0 var myFunx : MyFunx? while secondVar < 100 { print ("starting") if myFunx == nil { print ("making myFunx") DispatchQueue.global().async { myFunx = MyFunx() } } else { myFunx!.addValues(a: firstVar, b: secondVar) } firstVar += 1 secondVar += 1 } print ("myFunx = \(myFunx)") print ("\(myFunx?.opsCount)") print ("exiting")

I have my Swift app that records audio in chunks of multiple files, each M4A file is approx 1 minute long. I would like to go through those files and detect silence, or the lowest level. While I am able to read the file into a buffer, my problem is deciphering it. Even with Google, all it comes up with is "audio players" instead of sites that describe the header and the data. Where can I find what to look for? Or even if I should be reading it into a WAV file? But even then I cannot seem to find a tool, or a site, that tells me how to decipher what I am reading. Obviously it exists, since Siri knows when you've stopped speaking. Just trying to find the key.

Hello, Am starting to work with/learn the AVAudioEngine. Currently am at the point where I would like to be able read an audio file of a speech and determine if there are any moments of silence in the speech. Does this framework provide any such properties, such as power lever, decibels, etc. that I can use in finding long enough moments of silence?

I am trying to save the buffer from my installTap to a file. I do it in chunks of 10 so I'll get a bigger file. When I try to play the written file (from the simulator's directory) QuickTime says that it's not compatible. I have examined the bad m4a file and a working one. There are a lot of zero's in the bad file at the beginning followed by a lot of data. However both files appears to have the same header. A lot of people mention that I have to nil the AudioFile, but: audioFile = nil is not a valid syntax, nor can I file a close method in AudioFile. Here's the complete code, edited into one working file: import UIKit import AVFoundation class ViewController: UIViewController { let audioEngine = AVAudioEngine() var audioFile = AVAudioFile() var x = 0 override func viewDidLoad() { super.viewDidLoad() record() // Do any additional setup after loading the view. } func makeFile(format: AVAudioFormat) { let paths = FileManager.default.urls(for: .documentDirectory, in: .userDomainMask).first do { _ = try FileManager.default.contentsOfDirectory(at: paths!, includingPropertiesForKeys: nil) } catch { print ("error")} let destinationPath = paths!.appendingPathComponent("audioT.m4a") print ("\(destinationPath)") do { audioFile = try AVAudioFile(forWriting: destinationPath, settings: format.settings) print ("file created") } catch { print ("error creating file")} } func record(){ let node = audioEngine.inputNode let recordingFormat = node.inputFormat(forBus: 0) makeFile(format: recordingFormat) node.installTap(onBus: 0, bufferSize: 8192, format: recordingFormat, block: { [self] (buffer, _) in do { try audioFile.write(from: buffer); print ("buffer filled"); x += 1; print("wrote \(x)") if x > 9 { endThis() } } catch {return};}) audioEngine.prepare() do { try audioEngine.start() } catch let error { print ("oh catch") } } func endThis(){ audioEngine.stop() audioEngine.inputNode.removeTap(onBus: 0) } }

Am trying to go from the installTap straight to AVAudioFile(forWriting: I call: let recordingFormat = node.outputFormat(forBus: 0) and I get back : <AVAudioFormat 0x60000278f750:  1 ch,  48000 Hz, Float32> But AVAudioFile has a settings parameter of [String : Any] and am curious of how to place those values into recording the required format. Hopefully these are the values I need?

Expanding a speech to text demo, and while it works, I am still trying to learn Swift. Is .installTap the Swift version of a C callback function? From what I interpret here, every time the buffer becomes full, the code in between the last { } runs, as well, the code below it is also run. It almost feels like a callback combined with a GOTO line from basic. yes, it works, but I'd like to understand that I am getting the flow of the code correctly. func startSpeechRecognition (){ let node = audioEngine.inputNode let recordingFormat = node.outputFormat(forBus: 0) node.installTap(onBus: 0, bufferSize: 1024, format: recordingFormat) { (buffer, _) in self.request.append(buffer) } audioEngine.prepare() do { try audioEngine.start() } catch let error { ... } guard let myRecognition = SFSpeechRecognizer() else { ... return } if !myRecognition.isAvailable { ... } task = speechRecognizer?.recognitionTask(with: request, resultHandler: { (response, error) in guard let response = response else { if error != nil { print ("\(String(describing: error.debugDescription))") } else { print ("problem in repsonse") } return } let message = response.bestTranscription.formattedString print ("\(message)") }) }

I’m trying to do something really complex with audio streams. I.e. process the stream live edit it and then save it in snippets, all while the user is still speaking. I’m a book person, and reading hardcopy documentation is much easier for me.

Am using the demo code below to flesh out an audio recording app in Swift 5.x I would like to monitor certain aspects of the AVAudioRecorder as it is recording. Such as: frequency, power, volume, etc. but in live time. I found an example in Swift 3 where the user sets up a callback timer for 0.5 sec. I was wondering if this was still the case, or that in the latest version of Swift, there might be a callback function in the AVAudioEngine that gets called at a regular frequency? do { audioRecorder = try AVAudioRecorder(url: audioFilename!, settings: settings) audioRecorder.delegate = self audioRecorder.record() recordButton.setTitle("Tap to Stop", for: .normal) } catch { finishRecording(success: false) } }

I was under the impression, with offline speech to text, that there was no limit. Since the app wouldn't be using Apple's servers in real time. Yet when I process: speechRecognizer.recognitionTask it quits after one minute. Did I misread something ?

it's a great tool from Apple, but I want to delve more into its engine as I need to. The documentation doesn't seem to go there. For instance, I can't figure out how to clear the bestTranscritption object in speechRecognizer, as it always contains the entire transcription. There are other things I would like to work with as well. Has anyone worked with this heavily enough to recommend proper books are paid for tutorials? Many thanks

Building a very simple voice-to-text app, which I got from an online demo. What I can't seem to find is how to reset the response back to nil. This demo just keeps transcribing from the very beginning till it finally stalls. While I don't know how if the stall is related to my question, I still need to find out how to code "Ok, got the first 100 words. Reset response text to nil. Continue." func startSpeechRecognition(){ let node = audioEngine.inputNode let recordingFormat = node.outputFormat(forBus: 0) node.installTap(onBus: 0, bufferSize: 1024, format: recordingFormat, block: { (buffer, _) in self.request.append(buffer)}) audioEngine.prepare() do { try audioEngine.start() } catch let error { alertView(message: "audioEngine start error") } guard let myRecognition = SFSpeechRecognizer() else { self.alertView(message: "Recognition is not on your phone") return } if !myRecognition.isAvailable { self.alertView(message: "recognition is not available right now") } task = speechRecognizer?.recognitionTask(with: request, resultHandler: { (response, error) in guard let response = response else { if error != nil { self.alertView(message: error!.localizedDescription.debugDescription) } else { self.alertView(message: "Unknow error in creating task") } return } let message = response.bestTranscription.formattedString self.label.text = message }) }

Copied the code below from a tutorial, and I mostly understand what is going on. But I'd like to be able to fully read it. What I do get is that: handler can be nil, but here it's the code to run upon the completion of UIAlertAction. But am unsure what (_) in is. I have also sometimes seen it as [self] in Thank you controller.addAction(UIAlertAction(title: "OK", style: .default, handler: { (_)  in controller.dismiss(animated: true, completion: nil)}))

I've learned the hard way that specific commands to add a child UIView must be in a certain order, especially if I am bringing in the child UIView using animation. So I'd like to be clear that what the order I use to delete a child UIView is correct. Yes, what I have below works. But that doesn't mean it's correct. Thank you UIView.transition(with: parent, duration: 0.5, options: .transitionCurlUp, animations: { [self] in self.willMove(toParent: nil); self.removeFromParent(); self.view.removeFromSuperview(); self.dismiss(animated: false, completion: nil); }, completion:nil)

Thanks to people on this board I am able to successfully calla up a child UIViewConroller via animation with: This is the buttonAction from the Main UIViewController, which calls up setController @objc func buttonAction(sender: UIButton!) { guard let theButton = sender as? MyButton else { return} UIView.transition(with: self.view, duration: 0.5, options: .transitionCurlDown, animations: { [self] in self.addChild(setController); self.view.addSubview(setController.view); }, completion: { [self]_ in setController.didMove(toParent: self); setController.doLayout();}) } the doLayout method lies within the child: func doLayout (){ guard let parent = cView!.view.superview else {return} //make sure UIV honors safeAreaLayouts setConstraints(vc: self, pc: parent) } A button within the child, setController, dismisses itself: @objc func buttonAction(sender: UIButton!) { self.willMove(toParent: nil) self.removeFromParent() self.view.removeFromSuperview() self.dismiss(animated: false, completion: nil) } Everything works great the first time I call up the child UIView. It curls down while covering the first/parent UIVIEW, etc. etc. Figure 1 But after I dismiss the child view and call it again, the child view scrolls down without really covering the main view, it's like a mishmash. Figure 2 Only after all is said and done, then the child view covers everything. So am curious if I am dismissing something incorrectly.