我正在构建一个快速应用程序，允许用户对着手机说话并将转录内容保存到文本视图中。当我说出一个句子并转录它后，文本字段会被清除以打印下一个句子，问题就出现了。我希望它继续打印转录而不删除前面的句子。
希望获得有关此问题的一些帮助或见解！！
private var lastRecognizedText = ""
private var fullTranscriptionText = ""
private var isListening = false
private func startListening() {
if isListening {
print("Already listening, start request ignored.")
return
}
stopListening() // Stop any previous sessions
isListening = true
listeningAnimation?.startAnimating()
processedNotesAnimation?.startAnimating()
listeningLabel.isHidden = false
processedNotesLabel.isHidden = false
producedNotesView.isHidden = false
configureHeroModifiers()
styleTextView()
setupAudioSession()
recognitionRequest = SFSpeechAudioBufferRecognitionRequest()
guard let recognitionRequest = recognitionRequest else {
print("Failed to create recognition request.")
return
}
recognitionRequest.shouldReportPartialResults = true
let inputNode = audioEngine.inputNode
var currentSentenceText = "" // Temporary variable to hold intermediate sentence text
recognitionTask = speechRecognizer?.recognitionTask(with: recognitionRequest) { result, error in
if let result = result {
let newText = result.bestTranscription.formattedString
if newText != self.lastRecognizedText {
if result.isFinal {
// Append finalized text to fullTranscriptionText
self.fullTranscriptionText += newText + ". "
self.textView.text = self.fullTranscriptionText
// Print the final accumulated transcription to the console
print("Final Transcription: \(self.fullTranscriptionText)")
// Reset currentSentenceText after finalizing
currentSentenceText = ""
} else {
// Update intermediate sentence text for real-time correction
currentSentenceText = newText
// Display full transcription with the current intermediate text
self.textView.text = self.fullTranscriptionText + currentSentenceText
}
self.lastRecognizedText = newText
// Auto-scroll to show the latest additions
let bottom = NSMakeRange(self.textView.text.count - 1, 1)
self.textView.scrollRangeToVisible(bottom)
}
}
if let error = error {
print("Speech recognition error: \(error.localizedDescription)")
self.handleError(error)
self.stopListening() // Stop on error
}
}
let recordingFormat = inputNode.outputFormat(forBus: 0)
inputNode.installTap(onBus: 0, bufferSize: 1024, format: recordingFormat) { buffer, when in
self.recognitionRequest?.append(buffer)
}
audioEngine.prepare()
do {
try audioEngine.start()
print("Audio engine started.")
} catch {
print("Audio Engine Error: \(error.localizedDescription)")
}
textView.isScrollEnabled = true
textView.isEditable = false
}
@IBAction func stopRecordingTapped(_ sender: Any) {
stopListening()
}
private func stopListening() {
guard isListening else { return }
print("Stopping listening session.")
isListening = false
audioEngine.stop()
audioEngine.inputNode.removeTap(onBus: 0)
recognitionTask?.finish()
recognitionTask = nil
recognitionRequest = nil
}
private func setupAudioSession() {
let audioSession = AVAudioSession.sharedInstance()
do {
try audioSession.setCategory(.record, mode: .measurement, options: [.duckOthers, .allowBluetooth])
try audioSession.setActive(true, options: .notifyOthersOnDeactivation)
print("Audio session successfully configured.")
} catch {
print("Audio session setup error: \(error.localizedDescription)")
}
}
// Request permission for speech recognition and microphone
private func requestSpeechRecognitionPermission() {
SFSpeechRecognizer.requestAuthorization { authStatus in
OperationQueue.main.addOperation {
switch authStatus {
case .authorized:
print("Speech recognition authorized")
case .denied, .restricted, .notDetermined:
print("Speech recognition permission denied")
default:
break
}
}
}
AVAudioSession.sharedInstance().requestRecordPermission { granted in
if granted {
print("Microphone permission granted")
} else {
print("Microphone permission denied")
}
}
}
// Function to reset variables and clear old inputs
private func resetForNewSession() {
// Reset recognizedText to start fresh
recognizedText = ""
// Clear previous content from the transcript file
clearTranscriptFile()
}
// Function to clear the transcript file
private func clearTranscriptFile() {
do {
// Overwrite the file with an empty string
try "".write(to: transcriptFilePath, atomically: true, encoding: .utf8)
print("Transcript file cleared for new session")
} catch {
print("Error clearing transcript file: \(error)")
}
}
func setupAudioRecorder() {
let audioSession = AVAudioSession.sharedInstance()
// Configure audio session
do {
try audioSession.setCategory(.playAndRecord, mode: .default, options: .defaultToSpeaker)
try audioSession.setActive(true)
let settings: [String: Any] = [
AVFormatIDKey: Int(kAudioFormatAppleLossless),
AVSampleRateKey: 44100.0,
AVNumberOfChannelsKey: 1,
AVEncoderAudioQualityKey: AVAudioQuality.high.rawValue
]
audioRecorder = try AVAudioRecorder(url: URL(fileURLWithPath: "/dev/null"), settings: settings)
audioRecorder.isMeteringEnabled = true
audioRecorder.record()
} catch {
print("Failed to set up audio recorder: \(error)")
}
}
您忘记将每个
newTextfullTranscriptionTextresultrecognitionTask = speechRecognizer?.recognitionTask(with: recognitionRequest) { result, error in
if let result = result {
let newText = result.bestTranscription.formattedString
if newText != self.lastRecognizedText {
if result.isFinal {
// Append finalized text to fullTranscriptionText
self.fullTranscriptionText += newText + ". "
// self.textView.text = self.fullTranscriptionText
// Print the final accumulated transcription to the console
print("Final Transcription: \(self.fullTranscriptionText)")
// Reset currentSentenceText after finalizing
// currentSentenceText = ""
} else {
// Update intermediate sentence text for real-time correction
// currentSentenceText = newText
self.fullTranscriptionText += newText // <-- Add new to full text
// Display full transcription with the current intermediate text
// self.textView.text = self.fullTranscriptionText + currentSentenceText
}
self.lastRecognizedText = newText
// Auto-scroll to show the latest additions
// let bottom = NSMakeRange(self.textView.text.count - 1, 1)
// self.textView.scrollRangeToVisible(bottom)
}
}
if let error = error {
print("Speech recognition error: \(error.localizedDescription)")
self.speechError = error.localizedDescription
// self.handleError(error)
self.stopListening() // Stop on error
}