feat: avoid re-sending language detection event if last detected language or confidence don't change

Author: jamsch

android/src/main/java/expo/modules/speechrecognition/ExpoSpeechService.kt

@@ -60,6 +60,9 @@ class ExpoSpeechService(
     private var delayedFileStreamer: DelayedFileStreamer? = null
     private var soundState = SoundState.INACTIVE
 
+    private var lastDetectedLanguage: String? = null
+    private var lastLanguageConfidence: Float? = null
+
     var recognitionState = RecognitionState.INACTIVE
 
     companion object {
@@ -121,6 +124,8 @@ class ExpoSpeechService(
             audioRecorder = null
             delayedFileStreamer?.close()
             delayedFileStreamer = null
+            lastDetectedLanguage = null
+            lastLanguageConfidence = null
             recognitionState = RecognitionState.STARTING
             soundState = SoundState.INACTIVE
             lastVolumeChangeEventTime = 0L
@@ -435,7 +440,7 @@ class ExpoSpeechService(
         when {
             // File URI
             sourceUri.startsWith("file://") -> File(URI(sourceUri))
-            
+
             // Local file path without URI scheme
             !sourceUri.startsWith("https://") -> File(sourceUri)
 
@@ -624,11 +629,23 @@ class ExpoSpeechService(
     }
 
     override fun onLanguageDetection(results: Bundle) {
-        sendEvent("languagedetection", mapOf(
-            "detectedLanguage" to results.getString(SpeechRecognizer.DETECTED_LANGUAGE),
-            "confidence" to languageDetectionConfidenceLevelToFloat(results.getInt(SpeechRecognizer.LANGUAGE_DETECTION_CONFIDENCE_LEVEL)),
-            "topLocaleAlternatives" to results.getStringArrayList(SpeechRecognizer.TOP_LOCALE_ALTERNATIVES)
-        ))
+        val detectedLanguage = results.getString(SpeechRecognizer.DETECTED_LANGUAGE)
+        val confidence = languageDetectionConfidenceLevelToFloat(results.getInt(SpeechRecognizer.LANGUAGE_DETECTION_CONFIDENCE_LEVEL))
+
+        // Only send event if language or confidence has changed
+        if (detectedLanguage != lastDetectedLanguage || confidence != lastLanguageConfidence) {
+            lastDetectedLanguage = detectedLanguage
+            lastLanguageConfidence = confidence
+
+            sendEvent(
+                "languagedetection",
+                mapOf(
+                    "detectedLanguage" to detectedLanguage,
+                    "confidence" to confidence,
+                    "topLocaleAlternatives" to results.getStringArrayList(SpeechRecognizer.TOP_LOCALE_ALTERNATIVES),
+                ),
+            )
+        }
     }
 
     /**

example/App.tsx

@@ -93,8 +93,6 @@ export default function App() {
     const transcript = ev.results[0]?.transcript || "";
 
     setTranscription((current) => {
-      // When a final result comes in, we need to update the base transcript to build off from
-      // Because on Android and Web, multiple final results can be returned within a continuous session
       // When a final result is received, any following recognized transcripts will omit the previous final result
       const transcriptTally = ev.isFinal
         ? (current?.transcriptTally ?? "") + transcript
@@ -126,6 +124,10 @@ export default function App() {
     console.log("[event]: nomatch");
   });
 
+  useSpeechRecognitionEvent("languagedetection", (ev) => {
+    console.log("[event]: languagedetection", ev);
+  });
+
   const startListening = () => {
     if (status !== "idle") {
       return;
@@ -574,7 +576,12 @@ function GeneralSettings(props: {
                     : locale
                 }
                 active={settings.lang === locale}
-                onPress={() => handleChange("lang", locale)}
+                onPress={() =>
+                  handleChange(
+                    "lang",
+                    settings.lang === locale ? undefined : locale,
+                  )
+                }
               />
             );
           })}