react-native-kookit 0.3.7 → 0.3.8

package/ios/ReactNativeKookitModule.swift

@@ -1551,25 +1551,41 @@ public class ReactNativeKookitModule: Module {
 
     // TTS (Text-to-Speech) Function
     AsyncFunction("synthesizeSpeech") { (options: [String: Any], promise: Promise) in
-      Task {
-        do {
-          let text = options["text"] as? String ?? ""
-          if text.isEmpty {
-            promise.reject("TTS_ERROR", "Text is required")
-            return
-          }
-          
-          let voiceId = options["voiceId"] as? String
-          let language = options["language"] as? String ?? "en-US"
-          let pitch = options["pitch"] as? Double ?? 1.0
-          let rate = options["rate"] as? Double ?? 1.0
-          
-          // Initialize synthesizer if needed
+      // Extract parameters
+      let text = options["text"] as? String ?? ""
+      if text.isEmpty {
+        promise.reject("TTS_ERROR", "Text is required")
+        return
+      }
+      
+      let voiceId = options["voiceId"] as? String
+      let language = options["language"] as? String ?? "en-US"
+      let pitch = options["pitch"] as? Double ?? 1.0
+      let rate = options["rate"] as? Double ?? 1.0
+      //print the voiceId
+      print("voiceId: \(voiceId)")
+      //print the language
+      print("language: \(language)")
+      //print the pitch
+      print("pitch: \(pitch)")
+      //print the rate
+      print("rate: \(rate)")
+      
+      // Initialize synthesizer on main thread if needed
+      if self.speechSynthesizer == nil {
+        DispatchQueue.main.sync {
           if self.speechSynthesizer == nil {
-            self.speechSynthesizer = AVSpeechSynthesizer()
+            let synth = AVSpeechSynthesizer()
+            synth.usesApplicationAudioSession = true
+            self.speechSynthesizer = synth
           }
-          
-          let filePath = try await self.synthesizeSpeechToFile(
+        }
+      }
+      
+      // Perform synthesis on a background queue to avoid blocking
+      DispatchQueue.global(qos: .userInitiated).async {
+        do {
+          let filePath = try self.synthesizeSpeechToFileSync(
             text: text,
             voiceId: voiceId,
             language: language,
@@ -1686,78 +1702,126 @@ public class ReactNativeKookitModule: Module {
   
   // MARK: - TTS Helper Methods
   
-  private func synthesizeSpeechToFile(text: String, voiceId: String?, language: String, pitch: Float, rate: Float) async throws -> String {
-    return try await withCheckedThrowingContinuation { continuation in
-      guard let synthesizer = self.speechSynthesizer else {
-        continuation.resume(throwing: NSError(domain: "TTS", code: -1, userInfo: [NSLocalizedDescriptionKey: "Speech synthesizer not initialized"]))
-        return
-      }
-      
-      let utterance = AVSpeechUtterance(string: text)
-      
-      // Set voice
-      if let voiceId = voiceId {
-        utterance.voice = AVSpeechSynthesisVoice(identifier: voiceId)
+  // Synchronous version using write callback to generate audio file WITHOUT playing
+  private func synthesizeSpeechToFileSync(text: String, voiceId: String?, language: String, pitch: Float, rate: Float) throws -> String {
+    guard let synthesizer = self.speechSynthesizer else {
+      throw NSError(domain: "TTS", code: -1, userInfo: [NSLocalizedDescriptionKey: "Speech synthesizer not initialized"])
+    }
+    
+    // Create utterance
+    let utterance = AVSpeechUtterance(string: text)
+    
+    // Set voice
+    if let voiceId = voiceId {
+      utterance.voice = AVSpeechSynthesisVoice(identifier: voiceId)
+    } else {
+      let voices = AVSpeechSynthesisVoice.speechVoices()
+      if let voice = voices.first(where: { $0.language.hasPrefix(language) }) {
+        utterance.voice = voice
       } else {
-        // Try to find voice by language
-        let voices = AVSpeechSynthesisVoice.speechVoices()
-        if let voice = voices.first(where: { $0.language.hasPrefix(language) }) {
-          utterance.voice = voice
-        } else {
-          utterance.voice = AVSpeechSynthesisVoice(language: language)
-        }
+        utterance.voice = AVSpeechSynthesisVoice(language: language)
+      }
+    }
+    
+    // Set pitch and rate
+    utterance.pitchMultiplier = max(0.5, min(pitch, 2.0))
+    let minRate = AVSpeechUtteranceMinimumSpeechRate
+    let maxRate = AVSpeechUtteranceMaximumSpeechRate
+    let targetRate = rate * AVSpeechUtteranceDefaultSpeechRate
+    utterance.rate = min(max(targetRate, minRate), maxRate)
+    
+    // Create TTS cache directory
+    let cacheDir = FileManager.default.urls(for: .cachesDirectory, in: .userDomainMask)[0]
+    let ttsDir = cacheDir.appendingPathComponent("tts")
+    try FileManager.default.createDirectory(at: ttsDir, withIntermediateDirectories: true, attributes: nil)
+    
+    // Generate unique filename
+    let fileName = "tts_\(UUID().uuidString).caf"
+    let outputURL = ttsDir.appendingPathComponent(fileName)
+    
+    // Synchronization
+    let semaphore = DispatchSemaphore(value: 0)
+    var audioFile: AVAudioFile?
+    var capturedError: Error?
+    var bufferCount = 0
+    
+    print("🎤 Starting TTS audio generation (without playback)...")
+    
+    // Use write callback to generate audio WITHOUT playing
+    synthesizer.write(utterance) { buffer in
+      // Check for completion: buffer with frameLength == 0 OR buffer is nil
+      guard let pcmBuffer = buffer as? AVAudioPCMBuffer else {
+        // buffer is nil or not PCM - this means completion
+        print("✅ TTS generation completed, total buffers: \(bufferCount)")
+        semaphore.signal()
+        return
       }
       
-      // Set pitch and rate
-      utterance.pitchMultiplier = pitch
-      utterance.rate = rate * AVSpeechUtteranceDefaultSpeechRate
-      
-      // Create TTS cache directory
-      let cacheDir = FileManager.default.urls(for: .cachesDirectory, in: .userDomainMask)[0]
-      let ttsDir = cacheDir.appendingPathComponent("tts")
-      
-      do {
-        try FileManager.default.createDirectory(at: ttsDir, withIntermediateDirectories: true, attributes: nil)
-      } catch {
-        continuation.resume(throwing: error)
+      // Check if this is an empty buffer (completion signal)
+      if pcmBuffer.frameLength == 0 {
+        print("✅ TTS generation completed (empty buffer), total buffers: \(bufferCount)")
+        semaphore.signal()
         return
       }
       
-      // Generate unique filename
-      let fileName = "tts_\(UUID().uuidString).caf"
-      let outputURL = ttsDir.appendingPathComponent(fileName)
+      // Process non-empty buffer
+      bufferCount += 1
       
-      // Write utterance to file using AVSpeechSynthesizer.write
-      Task {
-        do {
-          var audioBuffers: [AVAudioPCMBuffer] = []
-          
-          synthesizer.write(utterance) { buffer in
-            if let buffer = buffer as? AVAudioPCMBuffer {
-              audioBuffers.append(buffer)
-            } else if buffer == nil {
-              // Writing is complete when buffer is nil
-              Task {
-                do {
-                  // Write all buffers to file
-                  try self.writeBuffersToFile(buffers: audioBuffers, url: outputURL)
-                  continuation.resume(returning: outputURL.path)
-                } catch {
-                  continuation.resume(throwing: error)
-                }
-              }
-            }
-          }
+      do {
+        // Create audio file on first buffer
+        if audioFile == nil {
+          print("📝 Creating audio file with format: \(pcmBuffer.format)")
+          audioFile = try AVAudioFile(
+            forWriting: outputURL,
+            settings: pcmBuffer.format.settings
+          )
+        }
+        
+        // Write buffer to file
+        try audioFile?.write(from: pcmBuffer)
+        
+        if bufferCount % 10 == 0 {
+          print("📦 Written \(bufferCount) buffers...")
         }
+      } catch {
+        print("❌ Error writing buffer: \(error)")
+        capturedError = error
+        semaphore.signal()
       }
     }
+    
+    // Wait for completion
+    print("⏳ Waiting for audio generation to complete...")
+    semaphore.wait()
+    
+    // Check for errors
+    if let error = capturedError {
+      throw error
+    }
+    
+    // Verify audio file was created
+    guard audioFile != nil else {
+      throw NSError(domain: "TTS", code: -1, userInfo: [NSLocalizedDescriptionKey: "No audio data was generated"])
+    }
+    
+    print("✅ Audio file created: \(outputURL.path)")
+    return outputURL.path
   }
   
   private func writeBuffersToFile(buffers: [AVAudioPCMBuffer], url: URL) throws {
+    guard !buffers.isEmpty else {
+      throw NSError(domain: "TTS", code: -1, userInfo: [NSLocalizedDescriptionKey: "No audio data generated"])
+    }
+    
     guard let firstBuffer = buffers.first else {
       throw NSError(domain: "TTS", code: -1, userInfo: [NSLocalizedDescriptionKey: "No audio data generated"])
     }
     
+    // Verify that the buffer has valid data
+    guard firstBuffer.frameLength > 0 else {
+      throw NSError(domain: "TTS", code: -1, userInfo: [NSLocalizedDescriptionKey: "Audio buffer is empty (frameLength is 0)"])
+    }
+    
     // Create audio file with settings from the buffer
     let settings: [String: Any] = [
       AVFormatIDKey: kAudioFormatLinearPCM,
@@ -1772,7 +1836,10 @@ public class ReactNativeKookitModule: Module {
     
     // Write all buffers to file
     for buffer in buffers {
-      try audioFile.write(from: buffer)
+      // Skip empty buffers
+      if buffer.frameLength > 0 {
+        try audioFile.write(from: buffer)
+      }
     }
   }
   

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "react-native-kookit",
-  "version": "0.3.7",
+  "version": "0.3.8",
   "description": "React Native module for intercepting volume button presses on iOS and Android, with FTP client functionality",
   "main": "build/index.js",
   "types": "build/index.d.ts",