diff --git a/AudioCap.xcodeproj/project.pbxproj b/AudioCap.xcodeproj/project.pbxproj
index 629b31c..471b220 100644
--- a/AudioCap.xcodeproj/project.pbxproj
+++ b/AudioCap.xcodeproj/project.pbxproj
@@ -7,34 +7,40 @@
 	objects = {
 
 /* Begin PBXBuildFile section */
+		CC537FC52E09E7CA00503A96 /* RealtimeAudioMonitor.swift in Sources */ = {isa = PBXBuildFile; fileRef = CC537FC42E09E7CA00503A96 /* RealtimeAudioMonitor.swift */; };
+		CC537FC72E09E7D500503A96 /* VoiceActivityDetector.swift in Sources */ = {isa = PBXBuildFile; fileRef = CC537FC62E09E7D500503A96 /* VoiceActivityDetector.swift */; };
+		CC98B1A62E17370900A4EDAB /* ProcessSelectionView.swift in Sources */ = {isa = PBXBuildFile; fileRef = CC98B1A02E17370900A4EDAB /* ProcessSelectionView.swift */; };
+		CC98B1A72E17370900A4EDAB /* RootView.swift in Sources */ = {isa = PBXBuildFile; fileRef = CC98B1A42E17370900A4EDAB /* RootView.swift */; };
+		CC98B1A82E17370900A4EDAB /* RecordingIndicator.swift in Sources */ = {isa = PBXBuildFile; fileRef = CC98B1A22E17370900A4EDAB /* RecordingIndicator.swift */; };
+		CC98B1A92E17370900A4EDAB /* FileProxyView.swift in Sources */ = {isa = PBXBuildFile; fileRef = CC98B19F2E17370900A4EDAB /* FileProxyView.swift */; };
+		CC98B1AA2E17370900A4EDAB /* RecordingView.swift in Sources */ = {isa = PBXBuildFile; fileRef = CC98B1A32E17370900A4EDAB /* RecordingView.swift */; };
+		CC98B1AB2E17370900A4EDAB /* RealtimeVADView.swift in Sources */ = {isa = PBXBuildFile; fileRef = CC98B1A12E17370900A4EDAB /* RealtimeVADView.swift */; };
 		F431728D2BF68C0C00D918A3 /* AudioRecordingPermission.swift in Sources */ = {isa = PBXBuildFile; fileRef = F431728C2BF68C0C00D918A3 /* AudioRecordingPermission.swift */; };
 		F43172902BF6A92900D918A3 /* AudioProcessController.swift in Sources */ = {isa = PBXBuildFile; fileRef = F431728F2BF6A92900D918A3 /* AudioProcessController.swift */; };
-		F43172922BF6ABB000D918A3 /* ProcessSelectionView.swift in Sources */ = {isa = PBXBuildFile; fileRef = F43172912BF6ABB000D918A3 /* ProcessSelectionView.swift */; };
 		F43172942BF6B01000D918A3 /* ProcessTap.swift in Sources */ = {isa = PBXBuildFile; fileRef = F43172932BF6B01000D918A3 /* ProcessTap.swift */; };
 		F43172962BF787C300D918A3 /* CoreAudioUtils.swift in Sources */ = {isa = PBXBuildFile; fileRef = F43172952BF787C300D918A3 /* CoreAudioUtils.swift */; };
-		F431729A2BF7A51A00D918A3 /* RecordingView.swift in Sources */ = {isa = PBXBuildFile; fileRef = F43172992BF7A51A00D918A3 /* RecordingView.swift */; };
-		F431729C2BF7B66A00D918A3 /* FileProxyView.swift in Sources */ = {isa = PBXBuildFile; fileRef = F431729B2BF7B66A00D918A3 /* FileProxyView.swift */; };
-		F43172A12BF7BF1300D918A3 /* RecordingIndicator.swift in Sources */ = {isa = PBXBuildFile; fileRef = F43172A02BF7BF1300D918A3 /* RecordingIndicator.swift */; };
 		F47AD9422BF5B61E005B75AC /* AudioCapApp.swift in Sources */ = {isa = PBXBuildFile; fileRef = F47AD9412BF5B61E005B75AC /* AudioCapApp.swift */; };
-		F47AD9442BF5B61E005B75AC /* RootView.swift in Sources */ = {isa = PBXBuildFile; fileRef = F47AD9432BF5B61E005B75AC /* RootView.swift */; };
 		F47AD9462BF5B61F005B75AC /* Assets.xcassets in Resources */ = {isa = PBXBuildFile; fileRef = F47AD9452BF5B61F005B75AC /* Assets.xcassets */; };
 		F47AD9492BF5B61F005B75AC /* Preview Assets.xcassets in Resources */ = {isa = PBXBuildFile; fileRef = F47AD9482BF5B61F005B75AC /* Preview Assets.xcassets */; };
 /* End PBXBuildFile section */
 
 /* Begin PBXFileReference section */
+		CC537FC42E09E7CA00503A96 /* RealtimeAudioMonitor.swift */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.swift; path = RealtimeAudioMonitor.swift; sourceTree = "<group>"; };
+		CC537FC62E09E7D500503A96 /* VoiceActivityDetector.swift */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.swift; path = VoiceActivityDetector.swift; sourceTree = "<group>"; };
+		CC98B19F2E17370900A4EDAB /* FileProxyView.swift */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.swift; path = FileProxyView.swift; sourceTree = "<group>"; };
+		CC98B1A02E17370900A4EDAB /* ProcessSelectionView.swift */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.swift; path = ProcessSelectionView.swift; sourceTree = "<group>"; };
+		CC98B1A12E17370900A4EDAB /* RealtimeVADView.swift */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.swift; path = RealtimeVADView.swift; sourceTree = "<group>"; };
+		CC98B1A22E17370900A4EDAB /* RecordingIndicator.swift */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.swift; path = RecordingIndicator.swift; sourceTree = "<group>"; };
+		CC98B1A32E17370900A4EDAB /* RecordingView.swift */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.swift; path = RecordingView.swift; sourceTree = "<group>"; };
+		CC98B1A42E17370900A4EDAB /* RootView.swift */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.swift; path = RootView.swift; sourceTree = "<group>"; };
 		F431728C2BF68C0C00D918A3 /* AudioRecordingPermission.swift */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.swift; path = AudioRecordingPermission.swift; sourceTree = "<group>"; };
 		F431728E2BF691D900D918A3 /* Info.plist */ = {isa = PBXFileReference; lastKnownFileType = text.plist; path = Info.plist; sourceTree = "<group>"; };
 		F431728F2BF6A92900D918A3 /* AudioProcessController.swift */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.swift; path = AudioProcessController.swift; sourceTree = "<group>"; };
-		F43172912BF6ABB000D918A3 /* ProcessSelectionView.swift */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.swift; path = ProcessSelectionView.swift; sourceTree = "<group>"; };
 		F43172932BF6B01000D918A3 /* ProcessTap.swift */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.swift; path = ProcessTap.swift; sourceTree = "<group>"; };
 		F43172952BF787C300D918A3 /* CoreAudioUtils.swift */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.swift; path = CoreAudioUtils.swift; sourceTree = "<group>"; };
-		F43172992BF7A51A00D918A3 /* RecordingView.swift */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.swift; path = RecordingView.swift; sourceTree = "<group>"; };
-		F431729B2BF7B66A00D918A3 /* FileProxyView.swift */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.swift; path = FileProxyView.swift; sourceTree = "<group>"; };
 		F431729E2BF7BB0E00D918A3 /* Main.xcconfig */ = {isa = PBXFileReference; lastKnownFileType = text.xcconfig; path = Main.xcconfig; sourceTree = "<group>"; };
-		F43172A02BF7BF1300D918A3 /* RecordingIndicator.swift */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.swift; path = RecordingIndicator.swift; sourceTree = "<group>"; };
 		F47AD93E2BF5B61E005B75AC /* AudioCap.app */ = {isa = PBXFileReference; explicitFileType = wrapper.application; includeInIndex = 0; path = AudioCap.app; sourceTree = BUILT_PRODUCTS_DIR; };
 		F47AD9412BF5B61E005B75AC /* AudioCapApp.swift */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.swift; path = AudioCapApp.swift; sourceTree = "<group>"; };
-		F47AD9432BF5B61E005B75AC /* RootView.swift */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.swift; path = RootView.swift; sourceTree = "<group>"; };
 		F47AD9452BF5B61F005B75AC /* Assets.xcassets */ = {isa = PBXFileReference; lastKnownFileType = folder.assetcatalog; path = Assets.xcassets; sourceTree = "<group>"; };
 		F47AD9482BF5B61F005B75AC /* Preview Assets.xcassets */ = {isa = PBXFileReference; lastKnownFileType = folder.assetcatalog; path = "Preview Assets.xcassets"; sourceTree = "<group>"; };
 		F47AD94A2BF5B61F005B75AC /* AudioCap.entitlements */ = {isa = PBXFileReference; lastKnownFileType = text.plist.entitlements; path = AudioCap.entitlements; sourceTree = "<group>"; };
@@ -51,6 +57,19 @@
 /* End PBXFrameworksBuildPhase section */
 
 /* Begin PBXGroup section */
+		CC98B1A52E17370900A4EDAB /* Views */ = {
+			isa = PBXGroup;
+			children = (
+				CC98B19F2E17370900A4EDAB /* FileProxyView.swift */,
+				CC98B1A02E17370900A4EDAB /* ProcessSelectionView.swift */,
+				CC98B1A12E17370900A4EDAB /* RealtimeVADView.swift */,
+				CC98B1A22E17370900A4EDAB /* RecordingIndicator.swift */,
+				CC98B1A32E17370900A4EDAB /* RecordingView.swift */,
+				CC98B1A42E17370900A4EDAB /* RootView.swift */,
+			);
+			path = Views;
+			sourceTree = "<group>";
+		};
 		F431729D2BF7BB0600D918A3 /* Config */ = {
 			isa = PBXGroup;
 			children = (
@@ -64,6 +83,8 @@
 			children = (
 				F431728C2BF68C0C00D918A3 /* AudioRecordingPermission.swift */,
 				F431728F2BF6A92900D918A3 /* AudioProcessController.swift */,
+				CC537FC42E09E7CA00503A96 /* RealtimeAudioMonitor.swift */,
+				CC537FC62E09E7D500503A96 /* VoiceActivityDetector.swift */,
 				F43172932BF6B01000D918A3 /* ProcessTap.swift */,
 				F43172952BF787C300D918A3 /* CoreAudioUtils.swift */,
 			);
@@ -92,11 +113,7 @@
 				F431729D2BF7BB0600D918A3 /* Config */,
 				F431729F2BF7BD4700D918A3 /* ProcessTap */,
 				F47AD9412BF5B61E005B75AC /* AudioCapApp.swift */,
-				F47AD9432BF5B61E005B75AC /* RootView.swift */,
-				F43172912BF6ABB000D918A3 /* ProcessSelectionView.swift */,
-				F43172992BF7A51A00D918A3 /* RecordingView.swift */,
-				F431729B2BF7B66A00D918A3 /* FileProxyView.swift */,
-				F43172A02BF7BF1300D918A3 /* RecordingIndicator.swift */,
+				CC98B1A52E17370900A4EDAB /* Views */,
 				F431728E2BF691D900D918A3 /* Info.plist */,
 				F47AD9452BF5B61F005B75AC /* Assets.xcassets */,
 				F47AD94A2BF5B61F005B75AC /* AudioCap.entitlements */,
@@ -141,7 +158,7 @@
 			attributes = {
 				BuildIndependentTargetsInParallel = 1;
 				LastSwiftUpdateCheck = 1530;
-				LastUpgradeCheck = 1530;
+				LastUpgradeCheck = 1640;
 				TargetAttributes = {
 					F47AD93D2BF5B61E005B75AC = {
 						CreatedOnToolsVersion = 15.3;
@@ -183,16 +200,19 @@
 			isa = PBXSourcesBuildPhase;
 			buildActionMask = 2147483647;
 			files = (
+				CC98B1A62E17370900A4EDAB /* ProcessSelectionView.swift in Sources */,
+				CC98B1A72E17370900A4EDAB /* RootView.swift in Sources */,
+				CC98B1A82E17370900A4EDAB /* RecordingIndicator.swift in Sources */,
+				CC98B1A92E17370900A4EDAB /* FileProxyView.swift in Sources */,
+				CC98B1AA2E17370900A4EDAB /* RecordingView.swift in Sources */,
+				CC98B1AB2E17370900A4EDAB /* RealtimeVADView.swift in Sources */,
+				CC537FC52E09E7CA00503A96 /* RealtimeAudioMonitor.swift in Sources */,
 				F43172902BF6A92900D918A3 /* AudioProcessController.swift in Sources */,
 				F43172942BF6B01000D918A3 /* ProcessTap.swift in Sources */,
-				F43172922BF6ABB000D918A3 /* ProcessSelectionView.swift in Sources */,
-				F47AD9442BF5B61E005B75AC /* RootView.swift in Sources */,
+				CC537FC72E09E7D500503A96 /* VoiceActivityDetector.swift in Sources */,
 				F47AD9422BF5B61E005B75AC /* AudioCapApp.swift in Sources */,
 				F431728D2BF68C0C00D918A3 /* AudioRecordingPermission.swift in Sources */,
-				F431729A2BF7A51A00D918A3 /* RecordingView.swift in Sources */,
-				F43172A12BF7BF1300D918A3 /* RecordingIndicator.swift in Sources */,
 				F43172962BF787C300D918A3 /* CoreAudioUtils.swift in Sources */,
-				F431729C2BF7B66A00D918A3 /* FileProxyView.swift in Sources */,
 			);
 			runOnlyForDeploymentPostprocessing = 0;
 		};
@@ -234,7 +254,9 @@
 				CLANG_WARN_UNREACHABLE_CODE = YES;
 				CLANG_WARN__DUPLICATE_METHOD_MATCH = YES;
 				COPY_PHASE_STRIP = NO;
+				DEAD_CODE_STRIPPING = YES;
 				DEBUG_INFORMATION_FORMAT = dwarf;
+				DEVELOPMENT_TEAM = BD72FKWLAY;
 				ENABLE_STRICT_OBJC_MSGSEND = YES;
 				ENABLE_TESTABILITY = YES;
 				ENABLE_USER_SCRIPT_SANDBOXING = YES;
@@ -299,7 +321,9 @@
 				CLANG_WARN_UNREACHABLE_CODE = YES;
 				CLANG_WARN__DUPLICATE_METHOD_MATCH = YES;
 				COPY_PHASE_STRIP = NO;
+				DEAD_CODE_STRIPPING = YES;
 				DEBUG_INFORMATION_FORMAT = "dwarf-with-dsym";
+				DEVELOPMENT_TEAM = BD72FKWLAY;
 				ENABLE_NS_ASSERTIONS = NO;
 				ENABLE_STRICT_OBJC_MSGSEND = YES;
 				ENABLE_USER_SCRIPT_SANDBOXING = YES;
@@ -328,8 +352,8 @@
 				CODE_SIGN_ENTITLEMENTS = AudioCap/AudioCap.entitlements;
 				CODE_SIGN_STYLE = Automatic;
 				COMBINE_HIDPI_IMAGES = YES;
+				DEAD_CODE_STRIPPING = YES;
 				DEVELOPMENT_ASSET_PATHS = "\"AudioCap/Preview Content\"";
-				DEVELOPMENT_TEAM = 8C7439RJLG;
 				ENABLE_HARDENED_RUNTIME = YES;
 				ENABLE_PREVIEWS = YES;
 				GENERATE_INFOPLIST_FILE = YES;
@@ -354,8 +378,8 @@
 				CODE_SIGN_ENTITLEMENTS = AudioCap/AudioCap.entitlements;
 				CODE_SIGN_STYLE = Automatic;
 				COMBINE_HIDPI_IMAGES = YES;
+				DEAD_CODE_STRIPPING = YES;
 				DEVELOPMENT_ASSET_PATHS = "\"AudioCap/Preview Content\"";
-				DEVELOPMENT_TEAM = 8C7439RJLG;
 				ENABLE_HARDENED_RUNTIME = YES;
 				ENABLE_PREVIEWS = YES;
 				GENERATE_INFOPLIST_FILE = YES;
diff --git a/AudioCap.xcodeproj/xcshareddata/xcschemes/AudioCap.xcscheme b/AudioCap.xcodeproj/xcshareddata/xcschemes/AudioCap.xcscheme
index c985a06..1f546fc 100644
--- a/AudioCap.xcodeproj/xcshareddata/xcschemes/AudioCap.xcscheme
+++ b/AudioCap.xcodeproj/xcshareddata/xcschemes/AudioCap.xcscheme
@@ -1,6 +1,6 @@
 <?xml version="1.0" encoding="UTF-8"?>
 <Scheme
-   LastUpgradeVersion = "1530"
+   LastUpgradeVersion = "1640"
    version = "1.7">
    <BuildAction
       parallelizeBuildables = "YES"
diff --git a/AudioCap/ProcessTap/RealtimeAudioMonitor.swift b/AudioCap/ProcessTap/RealtimeAudioMonitor.swift
new file mode 100644
index 0000000..1e71374
--- /dev/null
+++ b/AudioCap/ProcessTap/RealtimeAudioMonitor.swift
@@ -0,0 +1,117 @@
+import SwiftUI
+import AudioToolbox
+import AVFoundation
+import OSLog
+
+/// Real-time audio monitor for Voice Activity Detection
+@Observable
+final class RealtimeAudioMonitor {
+    
+    let process: AudioProcess
+    private let logger: Logger
+    private let vad = VoiceActivityDetector()
+    
+    @ObservationIgnored
+    private var tap: ProcessTap?
+    
+    private(set) var isMonitoring = false
+    private(set) var errorMessage: String?
+    
+    init(process: AudioProcess) {
+        self.process = process
+        self.logger = Logger(subsystem: kAppSubsystem, category: "\(String(describing: RealtimeAudioMonitor.self))(\(process.name))")
+    }
+    
+    /// Get the VAD instance for UI binding
+    var voiceActivityDetector: VoiceActivityDetector {
+        return vad
+    }
+    
+    /// Start real-time monitoring
+    @MainActor
+    func startMonitoring() throws {
+        guard !isMonitoring else {
+            logger.warning("Already monitoring")
+            return
+        }
+        
+        logger.debug("Starting real-time monitoring for \(self.process.name)")
+        
+        errorMessage = nil
+        
+        // Create a new tap for monitoring (separate from recording)
+        let monitorTap = ProcessTap(process: process, muteWhenRunning: false)
+        self.tap = monitorTap
+        
+        // Activate the tap
+        monitorTap.activate()
+        
+        if let error = monitorTap.errorMessage {
+            throw error
+        }
+        
+        // Start the monitoring process
+        let queue = DispatchQueue(label: "RealtimeAudioMonitor", qos: .userInitiated)
+        
+        guard let streamDescription = monitorTap.tapStreamDescription else {
+            throw "Tap stream description not available"
+        }
+        
+        var mutableStreamDescription = streamDescription
+        guard let format = AVAudioFormat(streamDescription: &mutableStreamDescription) else {
+            throw "Failed to create AVAudioFormat"
+        }
+        
+        logger.info("Monitoring with audio format: \(format)")
+        
+        try monitorTap.run(on: queue) { [weak self] inNow, inInputData, inInputTime, outOutputData, inOutputTime in
+            guard let self = self else { return }
+            
+            do {
+                // Create PCM buffer from input data
+                guard let buffer = AVAudioPCMBuffer(pcmFormat: format, bufferListNoCopy: inInputData, deallocator: nil) else {
+                    throw "Failed to create PCM buffer"
+                }
+                
+                // Process buffer through VAD (ensure this runs on main thread for @Observable updates)
+                DispatchQueue.main.async {
+                    self.vad.processAudioBuffer(buffer)
+                }
+                
+            } catch {
+                self.logger.error("Error processing audio buffer: \(error)")
+            }
+        } invalidationHandler: { [weak self] tap in
+            guard let self = self else { return }
+            DispatchQueue.main.async {
+                self.handleMonitoringStop()
+            }
+        }
+        
+        isMonitoring = true
+        logger.debug("Real-time monitoring started successfully")
+    }
+    
+    /// Stop real-time monitoring
+    @MainActor
+    func stopMonitoring() {
+        guard isMonitoring else { return }
+        
+        logger.debug("Stopping real-time monitoring")
+        
+        tap?.invalidate()
+        tap = nil
+        
+        handleMonitoringStop()
+    }
+    
+    private func handleMonitoringStop() {
+        isMonitoring = false
+        vad.reset()
+        logger.debug("Real-time monitoring stopped")
+    }
+    
+    deinit {
+        tap?.invalidate()
+    }
+} 
\ No newline at end of file
diff --git a/AudioCap/ProcessTap/VoiceActivityDetector.swift b/AudioCap/ProcessTap/VoiceActivityDetector.swift
new file mode 100644
index 0000000..53206eb
--- /dev/null
+++ b/AudioCap/ProcessTap/VoiceActivityDetector.swift
@@ -0,0 +1,153 @@
+import Foundation
+import AVFoundation
+import Accelerate
+import OSLog
+
+/// Represents a single audio frame analysis result
+struct AudioFrameAnalysis: Identifiable, Sendable {
+    let id = UUID()
+    let timestamp: TimeInterval
+    let rmsValue: Float
+    let isSpeech: Bool
+    let frameSize: Int
+    let sampleRate: Double
+    
+    var description: String {
+        "Frame: RMS=\(String(format: "%.4f", rmsValue)), Speech=\(isSpeech ? "YES" : "NO"), Size=\(frameSize)"
+    }
+}
+
+/// Voice Activity Detector using RMS analysis
+@Observable
+final class VoiceActivityDetector {
+    
+    private let logger = Logger(subsystem: kAppSubsystem, category: String(describing: VoiceActivityDetector.self))
+    
+    // VAD Configuration
+    private let rmsThreshold: Float = 0.01
+    private let targetFrameSize: Int = 4800 // For 48kHz sample rate (0.1 seconds)
+    
+    // Real-time analysis results
+    private(set) var recentFrames: [AudioFrameAnalysis] = []
+    private(set) var currentRMS: Float = 0.0
+    private(set) var isSpeechDetected: Bool = false
+    private(set) var totalFramesProcessed: Int = 0
+    private(set) var speechFrameCount: Int = 0
+    
+    // Audio buffer management
+    private var accumulatedSamples: [Float] = []
+    private var lastProcessTime = CFAbsoluteTimeGetCurrent()
+    
+    // Configuration
+    private let maxRecentFrames = 50 // Keep last 50 frames for UI display
+    
+    init() {
+        logger.debug("Initialized VAD with RMS threshold: \(self.rmsThreshold)")
+    }
+    
+    /// Process audio buffer and perform VAD analysis
+    func processAudioBuffer(_ buffer: AVAudioPCMBuffer) {
+        guard let floatChannelData = buffer.floatChannelData,
+              buffer.frameLength > 0 else {
+            logger.warning("Invalid audio buffer received")
+            return
+        }
+        
+        let frameLength = Int(buffer.frameLength)
+        let channelCount = Int(buffer.format.channelCount)
+        let sampleRate = buffer.format.sampleRate
+        
+        // Convert to mono if stereo (mix channels)
+        var monoSamples: [Float]
+        if channelCount == 1 {
+            monoSamples = Array(UnsafeBufferPointer(start: floatChannelData[0], count: frameLength))
+        } else {
+            // Mix stereo to mono
+            monoSamples = []
+            monoSamples.reserveCapacity(frameLength)
+            
+            for i in 0..<frameLength {
+                var sum: Float = 0.0
+                for channel in 0..<channelCount {
+                    sum += floatChannelData[channel][i]
+                }
+                monoSamples.append(sum / Float(channelCount))
+            }
+        }
+        
+        // Add samples to accumulation buffer
+        accumulatedSamples.append(contentsOf: monoSamples)
+        
+        // Process complete frames
+        let actualFrameSize = min(targetFrameSize, accumulatedSamples.count)
+        
+        while accumulatedSamples.count >= actualFrameSize {
+            let frameData = Array(accumulatedSamples.prefix(actualFrameSize))
+            accumulatedSamples.removeFirst(actualFrameSize)
+            
+            analyzeFrame(frameData, sampleRate: sampleRate)
+        }
+    }
+    
+    /// Analyze a single frame of audio data
+    private func analyzeFrame(_ samples: [Float], sampleRate: Double) {
+        let rms = calculateRMS(samples)
+        let isSpeech = rms > rmsThreshold
+        let timestamp = CFAbsoluteTimeGetCurrent()
+        
+        // Update current state
+        currentRMS = rms
+        isSpeechDetected = isSpeech
+        totalFramesProcessed += 1
+        
+        if isSpeech {
+            speechFrameCount += 1
+        }
+        
+        // Create analysis result
+        let analysis = AudioFrameAnalysis(
+            timestamp: timestamp,
+            rmsValue: rms,
+            isSpeech: isSpeech,
+            frameSize: samples.count,
+            sampleRate: sampleRate
+        )
+        
+        // Add to recent frames (keep only latest frames)
+        recentFrames.append(analysis)
+        if recentFrames.count > maxRecentFrames {
+            recentFrames.removeFirst()
+        }
+        
+        logger.debug("Frame analyzed: \(analysis.description)")
+    }
+    
+    /// Calculate RMS (Root Mean Square) of audio samples
+    private func calculateRMS(_ samples: [Float]) -> Float {
+        guard !samples.isEmpty else { return 0.0 }
+        
+        var rms: Float = 0.0
+        
+        // Use Accelerate framework for efficient computation
+        vDSP_rmsqv(samples, 1, &rms, vDSP_Length(samples.count))
+        
+        return rms
+    }
+    
+    /// Reset analysis state
+    func reset() {
+        accumulatedSamples.removeAll()
+        recentFrames.removeAll()
+        currentRMS = 0.0
+        isSpeechDetected = false
+        totalFramesProcessed = 0
+        speechFrameCount = 0
+        logger.debug("VAD state reset")
+    }
+    
+    /// Get speech activity percentage
+    var speechActivityPercentage: Float {
+        guard totalFramesProcessed > 0 else { return 0.0 }
+        return Float(speechFrameCount) / Float(totalFramesProcessed) * 100.0
+    }
+} 
\ No newline at end of file
diff --git a/AudioCap/FileProxyView.swift b/AudioCap/Views/FileProxyView.swift
similarity index 100%
rename from AudioCap/FileProxyView.swift
rename to AudioCap/Views/FileProxyView.swift
diff --git a/AudioCap/ProcessSelectionView.swift b/AudioCap/Views/ProcessSelectionView.swift
similarity index 52%
rename from AudioCap/ProcessSelectionView.swift
rename to AudioCap/Views/ProcessSelectionView.swift
index b9262c4..8e4667d 100644
--- a/AudioCap/ProcessSelectionView.swift
+++ b/AudioCap/Views/ProcessSelectionView.swift
@@ -5,6 +5,7 @@ struct ProcessSelectionView: View {
     @State private var processController = AudioProcessController()
     @State private var tap: ProcessTap?
     @State private var recorder: ProcessTapRecorder?
+    @State private var realtimeMonitor: RealtimeAudioMonitor?
 
     @State private var selectedProcess: AudioProcess?
 
@@ -39,8 +40,10 @@ struct ProcessSelectionView: View {
 
                 if let newValue {
                     setupRecording(for: newValue)
+                    setupRealtimeMonitoring(for: newValue)
                 } else if oldValue == tap?.process {
                     teardownTap()
+                    teardownRealtimeMonitoring()
                 }
             }
         } header: {
@@ -63,6 +66,69 @@ struct ProcessSelectionView: View {
                     }
             }
         }
+        
+        // Real-time VAD monitoring section
+        if let monitor = realtimeMonitor {
+            Section {
+                VStack(spacing: 12) {
+                    HStack {
+                        if monitor.isMonitoring {
+                            Button("Stop Real-time Analysis") {
+                                monitor.stopMonitoring()
+                            }
+                            .buttonStyle(.bordered)
+                        } else {
+                            Button("Start Real-time Analysis") {
+                                handlingErrors {
+                                    try monitor.startMonitoring()
+                                }
+                            }
+                            .buttonStyle(.borderedProminent)
+                        }
+                        
+                        Spacer()
+                        
+                        if monitor.isMonitoring {
+                            HStack {
+                                Circle()
+                                    .fill(Color.green)
+                                    .frame(width: 8, height: 8)
+                                    .opacity(0.8)
+                                    .animation(.easeInOut(duration: 1.0).repeatForever(autoreverses: true), value: monitor.isMonitoring)
+                                
+                                Text("Live")
+                                    .font(.caption)
+                                    .foregroundColor(.green)
+                            }
+                        }
+                    }
+                    
+                    if monitor.isMonitoring {
+                        RealtimeVADView(vad: monitor.voiceActivityDetector)
+                    } else {
+                        Text("Real-time voice activity detection analyzes audio in 4800-sample frames (0.1s at 48kHz) using RMS threshold of 0.01")
+                            .font(.caption)
+                            .foregroundColor(.secondary)
+                            .multilineTextAlignment(.center)
+                            .padding(.vertical, 8)
+                    }
+                }
+            } header: {
+                HStack {
+                    Image(systemName: "waveform.badge.magnifyingglass")
+                        .foregroundColor(.blue)
+                    Text("Real-time Analysis")
+                        .font(.headline)
+                }
+            }
+            
+            if let errorMessage = monitor.errorMessage {
+                Text(errorMessage)
+                    .font(.caption)
+                    .foregroundStyle(.red)
+                    .padding(.top, 4)
+            }
+        }
     }
 
     private func setupRecording(for process: AudioProcess) {
@@ -86,6 +152,24 @@ struct ProcessSelectionView: View {
     private func teardownTap() {
         tap = nil
     }
+    
+    private func setupRealtimeMonitoring(for process: AudioProcess) {
+        let monitor = RealtimeAudioMonitor(process: process)
+        self.realtimeMonitor = monitor
+    }
+    
+    private func teardownRealtimeMonitoring() {
+        realtimeMonitor?.stopMonitoring()
+        realtimeMonitor = nil
+    }
+    
+    private func handlingErrors(perform block: () throws -> Void) {
+        do {
+            try block()
+        } catch {
+            NSAlert(error: error).runModal()
+        }
+    }
 }
 
 extension URL {
diff --git a/AudioCap/Views/RealtimeVADView.swift b/AudioCap/Views/RealtimeVADView.swift
new file mode 100644
index 0000000..2cb161f
--- /dev/null
+++ b/AudioCap/Views/RealtimeVADView.swift
@@ -0,0 +1,174 @@
+import SwiftUI
+
+@MainActor
+struct RealtimeVADView: View {
+    let vad: VoiceActivityDetector
+    
+    var body: some View {
+        Section {
+            VStack(alignment: .leading, spacing: 12) {
+                // Current Status
+                HStack {
+                    Circle()
+                        .fill(vad.isSpeechDetected ? Color.green : Color.gray)
+                        .frame(width: 12, height: 12)
+                        .animation(.easeInOut(duration: 0.2), value: vad.isSpeechDetected)
+                    
+                    Text(vad.isSpeechDetected ? "Speech Detected" : "No Speech")
+                        .font(.headline)
+                        .foregroundColor(vad.isSpeechDetected ? .green : .secondary)
+                    
+                    Spacer()
+                    
+                    Text("RMS: \(String(format: "%.4f", vad.currentRMS))")
+                        .font(.caption)
+                        .monospaced()
+                        .foregroundColor(.secondary)
+                }
+                
+                // Statistics
+                HStack {
+                    VStack(alignment: .leading) {
+                        Text("Frames Processed")
+                            .font(.caption)
+                            .foregroundColor(.secondary)
+                        Text("\(vad.totalFramesProcessed)")
+                            .font(.title3)
+                            .fontWeight(.medium)
+                    }
+                    
+                    Spacer()
+                    
+                    VStack(alignment: .trailing) {
+                        Text("Speech Activity")
+                            .font(.caption)
+                            .foregroundColor(.secondary)
+                        Text("\(String(format: "%.1f", vad.speechActivityPercentage))%")
+                            .font(.title3)
+                            .fontWeight(.medium)
+                            .foregroundColor(vad.speechActivityPercentage > 0 ? .green : .secondary)
+                    }
+                }
+                .padding(.vertical, 4)
+                
+                // RMS Level Indicator
+                VStack(alignment: .leading, spacing: 4) {
+                    HStack {
+                        Text("RMS Level")
+                            .font(.caption)
+                            .foregroundColor(.secondary)
+                        
+                        Spacer()
+                        
+                        Text("Threshold: 0.01")
+                            .font(.caption)
+                            .foregroundColor(.secondary)
+                    }
+                    
+                    GeometryReader { geometry in
+                        ZStack(alignment: .leading) {
+                            // Background
+                            Rectangle()
+                                .fill(Color.gray.opacity(0.2))
+                                .frame(height: 8)
+                                .cornerRadius(4)
+                            
+                            // Threshold line
+                            Rectangle()
+                                .fill(Color.orange)
+                                .frame(width: 2, height: 12)
+                                .offset(x: geometry.size.width * (0.01 / 0.1)) // Assuming max scale of 0.1
+                            
+                            // Current level
+                            Rectangle()
+                                .fill(vad.isSpeechDetected ? Color.green : Color.blue)
+                                .frame(width: max(2, geometry.size.width * min(1.0, Double(vad.currentRMS) / 0.1)), height: 8)
+                                .cornerRadius(4)
+                                .animation(.easeOut(duration: 0.1), value: vad.currentRMS)
+                        }
+                    }
+                    .frame(height: 12)
+                }
+                
+                // Frame List
+                if !vad.recentFrames.isEmpty {
+                    Divider()
+                    
+                    VStack(alignment: .leading, spacing: 8) {
+                        Text("Recent Frames")
+                            .font(.caption)
+                            .foregroundColor(.secondary)
+                        
+                        ScrollView {
+                            LazyVStack(alignment: .leading, spacing: 2) {
+                                ForEach(vad.recentFrames.suffix(15).reversed()) { frame in
+                                    FrameAnalysisRow(frame: frame)
+                                }
+                            }
+                        }
+                        .frame(maxHeight: 200)
+                    }
+                }
+            }
+            .padding(.vertical, 4)
+        } header: {
+            HStack {
+                Image(systemName: "waveform.circle.fill")
+                    .foregroundColor(vad.isSpeechDetected ? .green : .secondary)
+                    .animation(.easeInOut(duration: 0.2), value: vad.isSpeechDetected)
+                
+                Text("Real-time Voice Activity Detection")
+                    .font(.headline)
+            }
+        }
+    }
+}
+
+struct FrameAnalysisRow: View {
+    let frame: AudioFrameAnalysis
+    
+    var body: some View {
+        HStack(spacing: 8) {
+            // Speech indicator
+            Circle()
+                .fill(frame.isSpeech ? Color.green : Color.gray.opacity(0.3))
+                .frame(width: 6, height: 6)
+            
+            // RMS value
+            Text(String(format: "%.4f", frame.rmsValue))
+                .font(.caption)
+                .monospaced()
+                .frame(width: 60, alignment: .trailing)
+                .foregroundColor(frame.isSpeech ? .primary : .secondary)
+            
+            // Speech status
+            Text(frame.isSpeech ? "SPEECH" : "SILENCE")
+                .font(.caption)
+                .fontWeight(frame.isSpeech ? .medium : .regular)
+                .frame(width: 60, alignment: .leading)
+                .foregroundColor(frame.isSpeech ? .green : .secondary)
+            
+            Spacer()
+            
+            // Frame info
+            Text("\(frame.frameSize) samples")
+                .font(.caption)
+                .foregroundColor(.secondary)
+        }
+        .padding(.horizontal, 4)
+        .padding(.vertical, 1)
+        .background(
+            RoundedRectangle(cornerRadius: 4)
+                .fill(frame.isSpeech ? Color.green.opacity(0.1) : Color.clear)
+        )
+    }
+}
+
+#if DEBUG
+#Preview {
+    Form {
+        RealtimeVADView(vad: VoiceActivityDetector())
+    }
+    .formStyle(.grouped)
+}
+#endif 
\ No newline at end of file
diff --git a/AudioCap/RecordingIndicator.swift b/AudioCap/Views/RecordingIndicator.swift
similarity index 100%
rename from AudioCap/RecordingIndicator.swift
rename to AudioCap/Views/RecordingIndicator.swift
diff --git a/AudioCap/RecordingView.swift b/AudioCap/Views/RecordingView.swift
similarity index 100%
rename from AudioCap/RecordingView.swift
rename to AudioCap/Views/RecordingView.swift
diff --git a/AudioCap/RootView.swift b/AudioCap/Views/RootView.swift
similarity index 100%
rename from AudioCap/RootView.swift
rename to AudioCap/Views/RootView.swift
diff --git a/README.md b/README.md
index 13fd28e..5ce566d 100644
--- a/README.md
+++ b/README.md
@@ -8,6 +8,16 @@ This project is provided as documentation for this new API to help developers of
 
 https://github.com/insidegui/AudioCap/assets/67184/95d72d1f-a4d6-4544-9d2f-a2ab99507cfc
 
+
+
+
+https://github.com/user-attachments/assets/efc03340-1d1c-46bc-92b4-ee6e1e763dbc
+
+
+
+
+
+
 ## API Description
 
 Here’s a brief summary of the new API added in macOS 14.4 and how to put everything together.