Wait for ios speechSynthesizer.speak to finish utterance - ios

Question

I have a list of strings that I need to display on the screen and simultaneously convert to speech using speechSynthesizer.speak
The issue I am facing is that I am not able how to figure out how to wait for one utterance to finish and them move to the next one. What happens is that the text is displayed one after the other without syncing with the audio.
func speechAndText(text: String){
let speechUtterance = AVSpeechUtterance(string: text)
speechUtterance.rate = AVSpeechUtteranceMaximumSpeechRate/2.0
speechUtterance.pitchMultiplier = 1.2
speechUtterance.volume = 1
speechUtterance.voice = AVSpeechSynthesisVoice(language: "en-GB")
DispatchQueue.main.async {
self.textView.text = text
}
self.speechSynthesizer.speak(speechUtterance)
}
func speakPara(_ phrases: [String]) {
if let phrase = phrases.first {
let group = DispatchGroup()
group.enter()
DispatchQueue.global(qos: .default).async {
self.speechAndText(text: phrase)
group.leave()
}
group.wait()
let rest = Array(phrases.dropFirst())
if !rest.isEmpty {
self.speakPara(rest)
}
}
}
The speakParafunction takes the list of strings to be uttered and sends one string at a time to speechAndText to convert to speech and write on the display.
Any help would be appreciated. Thank you.

Answer 1

No need for GCD, you can use the delegate methods of AVSpeechSynthesizer.
Set self as delegate:
self.speechSynthesizer.delegate = self
and implement didFinish by speaking the next text.
extension YourClass : AVSpeechSynthesizerDelegate {
func speechSynthesizer(_ synthesizer: AVSpeechSynthesizer,
didFinish utterance: AVSpeechUtterance) {
currentIndex += 1
if currentIndex < phrases.lastIndex {
self.speechAndText(text: phrases[currentIndex])
}
}
}
currentIndex and phrases are two instance properties. phrases stores the array of strings that you want to speak, and currentIndex stores the index of the text that is currently being spoken. speakPara can be implemented like so:
func speakPara(_ phrases: [String]) {
// reset both helper properties
self.phrases = phrases
self.currentIndex = 0
// stop speaking whatever it was speaking and start speaking the first item (if any)
self.speechSynthesizer.stopSpeaking(at: .immediate)
phrases.map { self.speechAndText(text: $0) }
}