Refactor Windows audio capture setup to streamline processing and remove redundant logging

package.json

@@ -1,7 +1,7 @@
 {
     "name": "cheating-daddy",
     "productName": "cheating-daddy",
-    "version": "0.5.6",
+    "version": "0.5.9",
     "description": "cheating daddy",
     "main": "src/index.js",
     "scripts": {
@@ -19,8 +19,8 @@
         "cheating daddy ai assistant for interviews"
     ],
     "author": {
-        "name": "sohzm",
+        "name": "ShiftyX1",
-        "email": "sohambharambe9@gmail.com"
+        "email": "lead@pyserve.org"
     },
     "license": "GPL-3.0",
     "dependencies": {

src/utils/openai-sdk.js

@@ -281,6 +281,7 @@ async function sendImageMessage(base64Image, prompt) {
 let audioChunks = [];
 let lastAudioTime = 0;
 const SILENCE_THRESHOLD_MS = 1500; // 1.5 seconds of silence
+let silenceCheckTimer = null;
 
 async function processAudioChunk(base64Audio, mimeType) {
     if (!openaiClient) {
@@ -294,8 +295,20 @@ async function processAudioChunk(base64Audio, mimeType) {
     audioChunks.push(buffer);
     lastAudioTime = now;
 
-    // Check for silence (no new audio for SILENCE_THRESHOLD_MS)
+    // Clear existing timer
-    // This is a simple approach - in production you'd want proper VAD
+    if (silenceCheckTimer) {
+        clearTimeout(silenceCheckTimer);
+    }
+
+    // Set timer to check for silence
+    silenceCheckTimer = setTimeout(async () => {
+        const silenceDuration = Date.now() - lastAudioTime;
+        if (silenceDuration >= SILENCE_THRESHOLD_MS && audioChunks.length > 0) {
+            console.log('Silence detected, flushing audio for transcription...');
+            await flushAudioAndTranscribe();
+        }
+    }, SILENCE_THRESHOLD_MS);
+
     return { success: true, buffering: true };
 }
 

src/utils/renderer.js

@@ -295,9 +295,6 @@ async function startCapture(screenshotIntervalSeconds = 5, imageQuality = 'mediu
             console.log('Linux capture started - system audio:', mediaStream.getAudioTracks().length > 0, 'microphone mode:', audioMode);
         } else {
             // Windows - use display media with loopback for system audio
-            logToMain('info', '=== Starting Windows audio capture ===');
-            cheatingDaddy.setStatus('Requesting screen & audio...');
-            
             mediaStream = await navigator.mediaDevices.getDisplayMedia({
                 video: {
                     frameRate: 1,
@@ -313,6 +310,7 @@ async function startCapture(screenshotIntervalSeconds = 5, imageQuality = 'mediu
                 },
             });
 
+            console.log('Windows capture started with loopback audio');
             const audioTracks = mediaStream.getAudioTracks();
             const videoTracks = mediaStream.getVideoTracks();
             
@@ -328,14 +326,8 @@ async function startCapture(screenshotIntervalSeconds = 5, imageQuality = 'mediu
                 })),
             });
 
-            if (audioTracks.length === 0) {
-                logToMain('warn', 'WARNING: No audio tracks! User must check "Share audio" in screen picker dialog');
-                cheatingDaddy.setStatus('Warning: No audio - enable "Share audio" checkbox');
-            } else {
-                logToMain('info', 'Audio track acquired, setting up processing...');
             // Setup audio processing for Windows loopback audio only
             setupWindowsLoopbackProcessing();
-            }
 
             if (audioMode === 'mic_only' || audioMode === 'both') {
                 let micStream = null;
@@ -451,38 +443,16 @@ function setupLinuxSystemAudioProcessing() {
 
 function setupWindowsLoopbackProcessing() {
     // Setup audio processing for Windows loopback audio only
-    logToMain('info', 'Setting up Windows loopback audio processing...');
-    
-    try {
     audioContext = new AudioContext({ sampleRate: SAMPLE_RATE });
-        
-        logToMain('info', 'AudioContext created:', {
-            state: audioContext.state,
-            sampleRate: audioContext.sampleRate,
-        });
-        
-        // Resume AudioContext if suspended (Chrome policy)
-        if (audioContext.state === 'suspended') {
-            logToMain('warn', 'AudioContext suspended, attempting resume...');
-            audioContext.resume().then(() => {
-                logToMain('info', 'AudioContext resumed successfully');
-            }).catch(err => {
-                logToMain('error', 'Failed to resume AudioContext:', err.message);
-            });
-        }
-        
     const source = audioContext.createMediaStreamSource(mediaStream);
     audioProcessor = audioContext.createScriptProcessor(BUFFER_SIZE, 1, 1);
 
     let audioBuffer = [];
     const samplesPerChunk = SAMPLE_RATE * AUDIO_CHUNK_DURATION;
-        let chunkCount = 0;
-        let totalSamples = 0;
 
     audioProcessor.onaudioprocess = async e => {
         const inputData = e.inputBuffer.getChannelData(0);
         audioBuffer.push(...inputData);
-            totalSamples += inputData.length;
 
         // Process audio in chunks
         while (audioBuffer.length >= samplesPerChunk) {
@@ -494,30 +464,11 @@ function setupWindowsLoopbackProcessing() {
                 data: base64Data,
                 mimeType: 'audio/pcm;rate=24000',
             });
-                
-                chunkCount++;
-                
-                // Log progress every 100 chunks (~10 seconds)
-                if (chunkCount === 1) {
-                    logToMain('info', 'First audio chunk sent to AI');
-                    cheatingDaddy.setStatus('Listening...');
-                } else if (chunkCount % 100 === 0) {
-                    // Calculate max amplitude to check if we're getting real audio
-                    const maxAmp = Math.max(...chunk.map(Math.abs));
-                    logToMain('info', `Audio progress: ${chunkCount} chunks, maxAmplitude: ${maxAmp.toFixed(4)}`);
-                }
         }
     };
 
     source.connect(audioProcessor);
     audioProcessor.connect(audioContext.destination);
-        
-        logToMain('info', 'Windows audio processing pipeline connected');
-        
-    } catch (err) {
-        logToMain('error', 'Error setting up Windows audio:', err.message, err.stack);
-        cheatingDaddy.setStatus('Audio error: ' + err.message);
-    }
 }
 
 async function captureScreenshot(imageQuality = 'medium', isManual = false) {

src/utils/window.js

@@ -33,14 +33,11 @@ function createWindow(sendToRenderer, geminiSessionRef) {
     });
 
     const { session, desktopCapturer } = require('electron');
-    
-    // Setup display media request handler for screen capture
-    // Use system picker on all platforms to let user choose screen/window and audio
     session.defaultSession.setDisplayMediaRequestHandler(
         (request, callback) => {
-            // Don't call callback - let system picker handle the selection
+            desktopCapturer.getSources({ types: ['screen'] }).then(sources => {
-            // The system will call callback with user's choice
+                callback({ video: sources[0], audio: 'loopback' });
-            console.log('Showing system screen picker dialog...');
+            });
         },
         { useSystemPicker: true }
     );