Merge pull request #152 from Ayush0Chaudhary/freemium

🧠 implemented the tap to stop panda, implement bring your own key

Author: Ayush0Chaudhary

app/build.gradle.kts

@@ -25,8 +25,9 @@ android {
         applicationId = "com.blurr.voice"
         minSdk = 24
         targetSdk = 35
-        versionCode = 8
-        versionName = "1.0.8"
+        versionCode = 9
+        versionName = "1.0.9" +
+                ""
 
         testInstrumentationRunner = "androidx.test.runner.AndroidJUnitRunner"
     }

app/src/main/assets/prompts/system_prompt.md

@@ -101,7 +101,7 @@ Strictly follow these rules while using the Android Phone and navigating the app
 - You have access to a persistent file system which you can use to track progress, store results, and manage long tasks.
 - Your file system is initialized with two files:
     1. `todo.md`: Use this to keep a checklist for known subtasks. Update it to mark completed items and track what remains. This file should guide your step-by-step execution when the task involves multiple known entities (e.g., a list of apps or items to visit). The contents of this file will be also visible in your state. ALWAYS use `write_file` to rewrite entire `todo.md` when you want to update your progress. NEVER use `append_file` on `todo.md` as this can explode your context.
-    2. `results.md`: Use this to accumulate extracted or generated results for the user. Append each new finding clearly and avoid duplication. This file serves as your output log.
+    2. `results.md`: Use this to accumulate extracted or generated results for the user. Append each new finding clearly and avoid duplication. This file serves as your output log but If user asked explicitly to summarize the screen, you will have to speak the summary using speak action, DONT JUST ADD THE RESULT, you are interacting with human too.
 - You can read, write, and append to files.
 - Note that `write_file` rewrites the entire file, so make sure to repeat all the existing information if you use this action.
 - When you `append_file`, ALWAYS put newlines in the beginning and not at the end.
@@ -127,7 +127,9 @@ The `done` action is your opportunity to terminate and share your findings with
 If you are allowed multiple actions:
 - You can specify multiple actions in the list to be executed sequentially (one after another). But always specify only one action name per item.
 - If the app-screen changes after an action, the sequence is interrupted and you get the new state. You might have to repeat the same action again so that your changes are reflected in the new state.
-- ONLY use multiple actions when actions should not change the screen state significantly.
+- ONLY use multiple actions when actions should not change the screen state significantly. 
+- If you think something needs to communicated with the user, please use speak command. For example request like summarize the current screen.
+- If user have question about the current screen, don't go to another app. 
 
 If you are allowed 1 action, ALWAYS output only 1 most reasonable action per step. If you have something in your read_state, always prioritize saving the data first.
 </action_rules>

app/src/main/java/com/blurr/voice/ConversationalAgentService.kt

@@ -98,6 +98,7 @@ class ConversationalAgentService : Service() {
         usedMemories.clear() // Clear used memories for new conversation
         visualFeedbackManager.showTtsWave()
         showInputBoxIfNeeded()
+        visualFeedbackManager.showSpeakingOverlay() // <-- ADD THIS LINE
 
 
     }
@@ -115,6 +116,11 @@ class ConversationalAgentService : Service() {
                 // This is the existing callback for when text is submitted
                 processUserInput(submittedText)
             },
+            onOutsideTap = {
+                serviceScope.launch {
+                    instantShutdown()
+                }
+            }
         )
     }
 
@@ -694,6 +700,9 @@ class ConversationalAgentService : Service() {
     }
 
     private suspend fun gracefulShutdown(exitMessage: String? = null) {
+        visualFeedbackManager.hideTtsWave()
+        visualFeedbackManager.hideTranscription()
+        visualFeedbackManager.hideSpeakingOverlay()
         visualFeedbackManager.hideInputBox()
 
         if (exitMessage != null) {
@@ -710,13 +719,40 @@ class ConversationalAgentService : Service() {
 
     }
 
+    /**
+     * Immediately stops all TTS, STT, and background tasks, hides all UI, and stops the service.
+     * This is used for forceful termination, like an outside tap.
+     */
+    private suspend fun instantShutdown() {
+        // We use the mainHandler to ensure these operations run on the UI thread.
+//        mainHandler.post {
+            Log.d("ConvAgent", "Instant shutdown triggered by user.")
+            speechCoordinator.stopSpeaking()
+            speechCoordinator.stopListening()
+            visualFeedbackManager.hideTtsWave()
+            visualFeedbackManager.hideTranscription()
+            visualFeedbackManager.hideSpeakingOverlay()
+            visualFeedbackManager.hideInputBox()
+
+            removeClarificationQuestions()
+            // Make a thread-safe copy of the conversation history.
+            if (conversationHistory.size > 1) {
+                Log.d("ConvAgent", "Extracting memories before shutdown.")
+                MemoryExtractor.extractAndStoreMemories(conversationHistory, memoryManager, usedMemories)
+            }
+            serviceScope.cancel("User tapped outside, forcing instant shutdown.")
+
+            stopSelf()
+//        }
+    }
     override fun onDestroy() {
         super.onDestroy()
         Log.d("ConvAgent", "Service onDestroy")
         removeClarificationQuestions()
         serviceScope.cancel()
         ttsManager.setCaptionsEnabled(false)
         isRunning = false
+        visualFeedbackManager.hideSpeakingOverlay() // <-- ADD THIS LINE
         // USE the new manager to hide the wave and transcription view
         visualFeedbackManager.hideTtsWave()
         visualFeedbackManager.hideTranscription()

app/src/main/java/com/blurr/voice/SettingsActivity.kt

@@ -15,6 +15,7 @@ import androidx.core.content.edit
 import androidx.lifecycle.lifecycleScope
 import com.blurr.voice.agent.v1.VisionMode
 import com.blurr.voice.api.GoogleTts
+import com.blurr.voice.api.PicovoiceKeyManager
 import com.blurr.voice.api.TTSVoice
 import com.blurr.voice.utilities.SpeechCoordinator
 import com.blurr.voice.utilities.VoicePreferenceManager
@@ -35,6 +36,8 @@ class SettingsActivity : AppCompatActivity() {
     private lateinit var visionModeDescription: TextView
     private lateinit var editUserName: android.widget.EditText
     private lateinit var editUserEmail: android.widget.EditText
+    private lateinit var editWakeWordKey: android.widget.EditText
+    private lateinit var buttonSaveWakeWordKey: Button
 
     private lateinit var sc: SpeechCoordinator
     private lateinit var sharedPreferences: SharedPreferences
@@ -78,6 +81,8 @@ class SettingsActivity : AppCompatActivity() {
         backButton = findViewById(R.id.id_backButtonSettings)
         permissionsInfoButton = findViewById(R.id.permissionsInfoButton)
         visionModeGroup = findViewById(R.id.visionModeGroup)
+        editWakeWordKey = findViewById(R.id.editWakeWordKey)
+        buttonSaveWakeWordKey = findViewById(R.id.buttonSaveWakeWordKey)
         visionModeDescription = findViewById(R.id.visionModeDescription)
         editUserName = findViewById(R.id.editUserName)
         editUserEmail = findViewById(R.id.editUserEmail)
@@ -107,6 +112,12 @@ class SettingsActivity : AppCompatActivity() {
             val intent = Intent(this, PermissionsActivity::class.java)
             startActivity(intent)
         }
+        buttonSaveWakeWordKey.setOnClickListener {
+            val userKey = editWakeWordKey.text.toString().trim()
+            val keyManager = PicovoiceKeyManager(this)
+            keyManager.saveUserProvidedKey(userKey) // You will create this method next
+            Toast.makeText(this, "Wake word key saved.", Toast.LENGTH_SHORT).show()
+        }
     }
 
     private fun setupAutoSavingListeners() {
@@ -192,6 +203,10 @@ class SettingsActivity : AppCompatActivity() {
     }
 
     private fun loadAllSettings() {
+
+        // Inside loadAllSettings()
+        val keyManager = PicovoiceKeyManager(this)
+        editWakeWordKey.setText(keyManager.getUserProvidedKey() ?: "") // You will create this method next
         val savedVoiceName = sharedPreferences.getString(KEY_SELECTED_VOICE, DEFAULT_VOICE.name)
         val savedVoice = availableVoices.find { it.name == savedVoiceName } ?: DEFAULT_VOICE
         ttsVoicePicker.value = availableVoices.indexOf(savedVoice)

app/src/main/java/com/blurr/voice/api/PicovoiceKeyManager.kt

@@ -26,7 +26,8 @@ class PicovoiceKeyManager(private val context: Context) {
         private const val TAG = "PicovoiceKeyManager"
         private const val PREFS_NAME = "PicovoicePrefs"
         private const val KEY_ACCESS_KEY = "access_key"
-        
+        // Inside companion object
+        private const val KEY_USER_PROVIDED_KEY = "user_provided_access_key"
         private const val API_URL = BuildConfig.GCLOUD_GATEWAY_URL
         private const val API_KEY_HEADER = "x-api-key"
         private const val API_KEY_VALUE = BuildConfig.GCLOUD_GATEWAY_PICOVOICE_KEY
@@ -44,6 +45,11 @@ class PicovoiceKeyManager(private val context: Context) {
      */
     suspend fun getAccessKey(): String? = withContext(Dispatchers.IO) {
         try {
+            val userKey = getUserProvidedKey()
+            if (!userKey.isNullOrBlank()) {
+                Log.d(TAG, "Using user-provided Picovoice access key")
+                return@withContext userKey
+            }
             // Check if we have a cached key
             val cachedKey = getCachedAccessKey()
             if (cachedKey != null) {
@@ -151,4 +157,20 @@ class PicovoiceKeyManager(private val context: Context) {
         }
         Log.d(TAG, "Cleared cached Picovoice access key")
     }
+
+    /**
+     * Saves a key provided by the user to SharedPreferences.
+     */
+    fun saveUserProvidedKey(accessKey: String) {
+        sharedPreferences.edit {
+            putString(KEY_USER_PROVIDED_KEY, accessKey)
+        }
+    }
+
+    /**
+     * Gets the key provided by the user from SharedPreferences.
+     */
+    fun getUserProvidedKey(): String? {
+        return sharedPreferences.getString(KEY_USER_PROVIDED_KEY, null)
+    }
 } 

app/src/main/java/com/blurr/voice/utilities/VisualFeedbackManager.kt

@@ -35,6 +35,7 @@ class VisualFeedbackManager private constructor(private val context: Context) {
     private var transcriptionView: TextView? = null
     private var inputBoxView: View? = null
 
+    private var speakingOverlay: View? = null
 
     companion object {
         private const val TAG = "VisualFeedbackManager"
@@ -73,6 +74,8 @@ class VisualFeedbackManager private constructor(private val context: Context) {
             ttsVisualizer?.stop()
             ttsVisualizer = null
             TTSManager.getInstance(context).utteranceListener = null
+            hideSpeakingOverlay()
+
             Log.d(TAG, "Audio wave effect has been torn down.")
         }
     }
@@ -124,6 +127,34 @@ class VisualFeedbackManager private constructor(private val context: Context) {
         Log.d(TAG, "Audio wave effect has been set up.")
     }
 
+    fun showSpeakingOverlay() {
+        mainHandler.post {
+            if (speakingOverlay != null) return@post
+
+            speakingOverlay = View(context).apply {
+                // CHANGED: Increased opacity from 80 (50%) to E6 (90%) for a more solid feel.
+                // You can adjust this hex value (E6) to your liking.
+                setBackgroundColor(0x80FFFFFF.toInt())
+            }
+
+            val params = WindowManager.LayoutParams(
+                WindowManager.LayoutParams.MATCH_PARENT,
+                WindowManager.LayoutParams.MATCH_PARENT,
+                WindowManager.LayoutParams.TYPE_APPLICATION_OVERLAY,
+                WindowManager.LayoutParams.FLAG_NOT_FOCUSABLE or WindowManager.LayoutParams.FLAG_NOT_TOUCHABLE,
+                PixelFormat.TRANSLUCENT
+            )
+
+            try {
+                windowManager.addView(speakingOverlay, params)
+                Log.d(TAG, "Speaking overlay added.")
+            } catch (e: Exception) {
+                Log.e(TAG, "Error adding speaking overlay", e)
+            }
+        }
+    }
+
+
     fun showTranscription(initialText: String = "Listening...") {
         if (transcriptionView != null) {
             updateTranscription(initialText) // Update text if already shown
@@ -193,7 +224,8 @@ class VisualFeedbackManager private constructor(private val context: Context) {
     @SuppressLint("ClickableViewAccessibility")
     fun showInputBox(
         onActivated: () -> Unit,
-        onSubmit: (String) -> Unit
+        onSubmit: (String) -> Unit,
+        onOutsideTap: () -> Unit
     ) {
         mainHandler.post {
             if (inputBoxView?.isAttachedToWindow == true) {
@@ -217,9 +249,9 @@ class VisualFeedbackManager private constructor(private val context: Context) {
             val params = WindowManager.LayoutParams(
                 WindowManager.LayoutParams.MATCH_PARENT, WindowManager.LayoutParams.WRAP_CONTENT,
                 WindowManager.LayoutParams.TYPE_APPLICATION_OVERLAY,
-                // You need it to be focusable to receive key events.
-                // The absence of FLAG_NOT_FOCUSABLE makes it focusable by default.
-                WindowManager.LayoutParams.FLAG_LAYOUT_IN_SCREEN or WindowManager.LayoutParams.FLAG_WATCH_OUTSIDE_TOUCH,
+                WindowManager.LayoutParams.FLAG_LAYOUT_IN_SCREEN or
+                        WindowManager.LayoutParams.FLAG_WATCH_OUTSIDE_TOUCH or
+                        WindowManager.LayoutParams.FLAG_NOT_TOUCH_MODAL,
                 PixelFormat.TRANSLUCENT
             ).apply {
                 gravity = Gravity.BOTTOM
@@ -264,7 +296,8 @@ class VisualFeedbackManager private constructor(private val context: Context) {
 
             rootLayout?.setOnTouchListener { _, event ->
                 if (event.action == MotionEvent.ACTION_OUTSIDE) {
-                    hideInputBox()
+                    Log.d(TAG, "Outside touch detected.")
+                    onOutsideTap() // Use the new callback
                     return@setOnTouchListener true
                 }
                 false
@@ -295,4 +328,19 @@ class VisualFeedbackManager private constructor(private val context: Context) {
             inputBoxView = null
         }
     }
+    fun hideSpeakingOverlay() {
+        mainHandler.post {
+            speakingOverlay?.let {
+                if (it.isAttachedToWindow) {
+                    try {
+                        windowManager.removeView(it)
+                        Log.d(TAG, "Speaking overlay removed.")
+                    } catch (e: Exception) {
+                        Log.e(TAG, "Error removing speaking overlay", e)
+                    }
+                }
+            }
+            speakingOverlay = null
+        }
+    }
 }

app/src/main/res/drawable/edit_text_border.xml

@@ -0,0 +1,11 @@
+<?xml version="1.0" encoding="utf-8"?>
+<shape xmlns:android="http://schemas.android.com/apk/res/android"
+    android:shape="rectangle">
+
+    <solid android:color="#333333" />
+
+    <stroke
+        android:width="1dp"
+        android:color="#888888" /> <corners android:radius="12dp" />
+
+</shape>

app/src/main/res/layout/activity_settings.xml

@@ -85,6 +85,53 @@
             </FrameLayout>
 
 
+        </LinearLayout>
+        <LinearLayout
+            android:layout_width="match_parent"
+            android:layout_height="wrap_content"
+            android:layout_marginTop="16dp"
+            android:background="@drawable/rounded_background"
+            android:orientation="vertical"
+            android:padding="16dp">
+
+            <TextView
+                android:layout_width="wrap_content"
+                android:layout_height="wrap_content"
+                android:text="Wake Word Service Key"
+                android:textColor="@color/white"
+                android:textSize="18sp"
+                android:textStyle="bold" />
+
+            <TextView
+                android:layout_width="wrap_content"
+                android:layout_height="wrap_content"
+                android:layout_marginTop="4dp"
+                android:text="If you have your own Picovoice AccessKey, you can paste it here."
+                android:textColor="#888888"
+                android:textSize="12sp" />
+
+            <EditText
+                android:id="@+id/editWakeWordKey"
+                android:layout_width="match_parent"
+                android:layout_height="wrap_content"
+                android:layout_marginTop="12dp"
+                android:background="@drawable/edit_text_border"
+                android:hint="Paste your key here"
+                android:inputType="text"
+                android:padding="12dp"
+                android:textColor="#CECECE"
+                android:textColorHint="#888888" />
+
+            <Button
+                android:id="@+id/buttonSaveWakeWordKey"
+                android:layout_width="wrap_content"
+                android:layout_height="wrap_content"
+                android:layout_gravity="end"
+                android:layout_marginTop="12dp"
+                android:text="Save Key"
+                android:background="@drawable/btn_with_border"
+                android:textColor="@android:color/white" />
+
         </LinearLayout>
 
         <LinearLayout
@@ -203,7 +250,6 @@
         </LinearLayout>
 
 
-
         <TextView
             android:id="@+id/permissionsInfoButton"
             android:layout_width="wrap_content"