Support Queue Strategy on Android and iOS

README.md

@@ -51,6 +51,7 @@ const speak = async () => {
     pitch: 1.0,
     volume: 1.0,
     category: 'ambient',
+    queueStrategy: 1
   });
 };
 
@@ -201,6 +202,7 @@ addListener(eventName: 'onRangeStart', listenerFunc: (info: { start: number; end
 | **`volume`**   | <code>number</code> | The volume that the utterance will be spoken at.                                                                                                                               | <code>1.0</code>       |
 | **`voice`**    | <code>number</code> | The index of the selected voice that will be used to speak the utterance. Possible voices can be queried using `getSupportedVoices`.                                           |                        |
 | **`category`** | <code>string</code> | Select the iOS Audio session category. Possible values: `ambient` and `playback`. Use `playback` to play audio even when the app is in the background. Only available for iOS. | <code>"ambient"</code> |
+| **`queueStrategy`** | <code>number</code> | Select the strategy to adopt when several requests to speak overlap. Possible values: `0` and `1`. Use `0` to stop the current request when a new request is sent. Use `1` to buffer the speech request. The request will be executed when all previous requests have been completed. | <code>0</code> |
 
 
 #### SpeechSynthesisVoice

android/src/main/java/com/getcapacitor/community/tts/SpeakResultCallback.java

@@ -3,5 +3,5 @@
 public interface SpeakResultCallback {
     void onDone();
     void onError();
-    void onRangeStart(int start, int end, String spokenWord);
+    void onRangeStart(int start, int end);
 }

android/src/main/java/com/getcapacitor/community/tts/TextToSpeech.java

@@ -15,6 +15,7 @@
 import java.util.Collections;
 import java.util.HashMap;
 import java.util.Locale;
+import java.util.Map;
 import java.util.Set;
 
 public class TextToSpeech implements android.speech.tts.TextToSpeech.OnInitListener {
@@ -25,11 +26,44 @@ public class TextToSpeech implements android.speech.tts.TextToSpeech.OnInitListe
     private android.speech.tts.TextToSpeech tts = null;
     private int initializationStatus;
     private JSObject[] supportedVoices = null;
+    private Map<String, SpeakResultCallback> requests = new HashMap();
 
     TextToSpeech(Context context) {
         this.context = context;
         try {
             tts = new android.speech.tts.TextToSpeech(context, this);
+            tts.setOnUtteranceProgressListener(
+                    new UtteranceProgressListener() {
+                        @Override
+                        public void onStart(String utteranceId) {}
+
+                        @Override
+                        public void onDone(String utteranceId) {
+                            SpeakResultCallback callback = requests.get(utteranceId);
+                            if(callback != null) {
+                                callback.onDone();
+                                requests.remove(utteranceId);
+                            }
+                        }
+
+                        @Override
+                        public void onError(String utteranceId) {
+                            SpeakResultCallback callback = requests.get(utteranceId);
+                            if(callback != null) {
+                                callback.onError();
+                                requests.remove(utteranceId);
+                            }
+                        }
+
+                        @Override
+                        public void onRangeStart(String utteranceId, int start, int end, int frame) {
+                            SpeakResultCallback callback = requests.get(utteranceId);
+                            if(callback != null) {
+                                callback.onRangeStart(start, end);
+                            }
+                        }
+                    }
+            );
         } catch (Exception ex) {
             Log.d(LOG_TAG, ex.getLocalizedMessage());
         }
@@ -40,6 +74,19 @@ public void onInit(int status) {
         this.initializationStatus = status;
     }
 
+    public void speak(
+            String text,
+            String lang,
+            float rate,
+            float pitch,
+            float volume,
+            int voice,
+            String callbackId,
+            SpeakResultCallback resultCallback
+    ) {
+        speak(text, lang, rate, pitch, volume, voice, callbackId, resultCallback, android.speech.tts.TextToSpeech.QUEUE_FLUSH);
+    }
+
     public void speak(
         String text,
         String lang,
@@ -48,31 +95,13 @@ public void speak(
         float volume,
         int voice,
         String callbackId,
-        SpeakResultCallback resultCallback
+        SpeakResultCallback resultCallback,
+        int queueStrategy
     ) {
-        tts.stop();
-        tts.setOnUtteranceProgressListener(
-            new UtteranceProgressListener() {
-                @Override
-                public void onStart(String utteranceId) {}
-
-                @Override
-                public void onDone(String utteranceId) {
-                    resultCallback.onDone();
-                }
-
-                @Override
-                public void onError(String utteranceId) {
-                    resultCallback.onError();
-                }
-
-                @Override
-                public void onRangeStart(String utteranceId, int start, int end, int frame) {
-                    String spokenWord = text.substring(start, end);
-                    resultCallback.onRangeStart(start, end, spokenWord);
-                }
-            }
-        );
+        if(queueStrategy != android.speech.tts.TextToSpeech.QUEUE_ADD) {
+            stop();
+        }
+        requests.put(callbackId, resultCallback);
 
         Locale locale = Locale.forLanguageTag(lang);
 
@@ -92,8 +121,7 @@ public void onRangeStart(String utteranceId, int start, int end, int frame) {
                     int resultCode = tts.setVoice(newVoice);
                 }
             }
-
-            tts.speak(text, android.speech.tts.TextToSpeech.QUEUE_FLUSH, ttsParams, callbackId);
+            tts.speak(text, queueStrategy, ttsParams, callbackId);
         } else {
             HashMap<String, String> ttsParams = new HashMap<>();
             ttsParams.put(android.speech.tts.TextToSpeech.Engine.KEY_PARAM_UTTERANCE_ID, callbackId);
@@ -102,12 +130,13 @@ public void onRangeStart(String utteranceId, int start, int end, int frame) {
             tts.setLanguage(locale);
             tts.setSpeechRate(rate);
             tts.setPitch(pitch);
-            tts.speak(text, android.speech.tts.TextToSpeech.QUEUE_FLUSH, ttsParams);
+            tts.speak(text, queueStrategy, ttsParams);
         }
     }
 
     public void stop() {
         tts.stop();
+        requests.clear();
     }
 
     public JSArray getSupportedLanguages() {

android/src/main/java/com/getcapacitor/community/tts/TextToSpeechPlugin.java

@@ -1,6 +1,8 @@
 package com.getcapacitor.community.tts;
 
 import android.util.Base64;
+import android.util.Log;
+
 import com.getcapacitor.JSArray;
 import com.getcapacitor.JSObject;
 import com.getcapacitor.Plugin;
@@ -37,6 +39,7 @@ public void speak(PluginCall call) {
         float pitch = call.getFloat("pitch", 1.0f);
         float volume = call.getFloat("volume", 1.0f);
         int voice = call.getInt("voice", -1);
+        int queueStrategy = call.getInt("queueStrategy", 0);
 
         boolean isLanguageSupported = implementation.isLanguageSupported(lang);
         if (!isLanguageSupported) {
@@ -56,17 +59,18 @@ public void onError() {
             }
 
             @Override
-            public void onRangeStart(int start, int end, String spokenWord) {
+            public void onRangeStart(int start, int end) {
                 JSObject ret = new JSObject();
                 ret.put("start", start);
                 ret.put("end", end);
+                String spokenWord = text.substring(start, end);
                 ret.put("spokenWord", spokenWord);
                 notifyListeners("onRangeStart", ret);
             }
         };
 
         try {
-            implementation.speak(text, lang, rate, pitch, volume, voice, call.getCallbackId(), resultCallback);
+            implementation.speak(text, lang, rate, pitch, volume, voice, call.getCallbackId(), resultCallback, queueStrategy);
         } catch (Exception ex) {
             call.reject(ex.getLocalizedMessage());
         }

ios/Plugin/TextToSpeech.swift

@@ -1,6 +1,10 @@
 import AVFoundation
 import Capacitor
 
+enum QUEUE_STRATEGY: Int {
+    case QUEUE_ADD = 1, QUEUE_FLUSH = 0
+}
+
 @objc public class TextToSpeech: NSObject, AVSpeechSynthesizerDelegate {
     let synthesizer = AVSpeechSynthesizer()
     var calls: [CAPPluginCall] = []
@@ -29,8 +33,10 @@ import Capacitor
         self.resolveCurrentCall()
     }
 
-    @objc public func speak(_ text: String, _ lang: String, _ rate: Float, _ pitch: Float, _ category: String, _ volume: Float, _ voice: Int, _ call: CAPPluginCall) throws {
-        self.synthesizer.stopSpeaking(at: .immediate)
+    @objc public func speak(_ text: String, _ lang: String, _ rate: Float, _ pitch: Float, _ category: String, _ volume: Float, _ voice: Int, _ queueStrategy: Int, _ call: CAPPluginCall) throws {
+        if(queueStrategy == QUEUE_STRATEGY.QUEUE_FLUSH.rawValue) {
+            self.synthesizer.stopSpeaking(at: .immediate)
+        }
         self.calls.append(call)
 
         let utterance = AVSpeechUtterance(string: text)

ios/Plugin/TextToSpeechPlugin.swift

@@ -20,6 +20,7 @@ public class TextToSpeechPlugin: CAPPlugin {
         let volume = call.getFloat("volume") ?? 1.0
         let voice = call.getInt("voice") ?? -1
         let category = call.getString("category") ?? "ambient"
+        let queueStrategy = call.getInt("queueStrategy") ?? 0
 
         let isLanguageSupported = implementation.isLanguageSupported(lang)
         guard isLanguageSupported else {
@@ -28,7 +29,7 @@ public class TextToSpeechPlugin: CAPPlugin {
         }
 
         do {
-            try implementation.speak(text, lang, rate, pitch, category, volume, voice, call)
+            try implementation.speak(text, lang, rate, pitch, category, volume, voice, queueStrategy, call)
         } catch {
             call.reject(error.localizedDescription)
         }

src/definitions.ts

@@ -87,6 +87,15 @@ export interface TTSOptions {
    * @default "ambient"
    */
   category?: string;
+  /**
+   * Select the strategy to adopt when several requests to speak overlap. 
+   * Possible values: `0` and `1`. 
+   * Use `0` to stop the current request when a new request is sent. 
+   * Use `1` to buffer the speech request. The request will be executed when all previous requests have been completed.
+   * 
+   * @default 0
+   */
+  queueStrategy?: number;
 }
 
 /**