diff --git a/VoiceInk/Whisper/LibWhisper.swift b/VoiceInk/Whisper/LibWhisper.swift
index 622d598..2703152 100644
--- a/VoiceInk/Whisper/LibWhisper.swift
+++ b/VoiceInk/Whisper/LibWhisper.swift
@@ -18,9 +18,7 @@ actor WhisperContext {
     private var promptCString: [CChar]?
     private let logger = Logger(subsystem: "com.prakashjoshipax.voiceink", category: "WhisperContext")
 
-    private init() {
-        // Private initializer without context
-    }
+    private init() {}
 
     init(context: OpaquePointer) {
         self.context = context
@@ -35,22 +33,18 @@ actor WhisperContext {
     func fullTranscribe(samples: [Float]) async {
         guard let context = context else { return }
         
-        // Leave 2 processors free (i.e. the high-efficiency cores).
         let maxThreads = max(1, min(8, cpuCount() - 2))
         var params = whisper_full_default_params(WHISPER_SAMPLING_GREEDY)
         
-        // Read language directly from UserDefaults
         let selectedLanguage = UserDefaults.standard.string(forKey: "SelectedLanguage") ?? "auto"
         if selectedLanguage != "auto" {
             languageCString = Array(selectedLanguage.utf8CString)
             params.language = languageCString?.withUnsafeBufferPointer { ptr in
                 ptr.baseAddress
             }
-            logger.notice("🌐 Using language: \(selectedLanguage)")
         } else {
             languageCString = nil
             params.language = nil
-            logger.notice("🌐 Using auto language detection")
         }
         
         if prompt != nil {
@@ -58,53 +52,46 @@ actor WhisperContext {
             params.initial_prompt = promptCString?.withUnsafeBufferPointer { ptr in
                 ptr.baseAddress
             }
-            logger.notice("💬 Using prompt for transcription in language: \(selectedLanguage)")
         } else {
             promptCString = nil
             params.initial_prompt = nil
         }
         
-        params.print_realtime   = true
-        params.print_progress   = false
+        params.print_realtime = true
+        params.print_progress = false
         params.print_timestamps = true
-        params.print_special    = false
-        params.translate        = false
-        params.n_threads        = Int32(maxThreads)
-        params.offset_ms        = 0
-        params.no_context       = true
-        params.single_segment   = false
+        params.print_special = false
+        params.translate = false
+        params.n_threads = Int32(maxThreads)
+        params.offset_ms = 0
+        params.no_context = true
+        params.single_segment = false
+        params.suppress_nst = true
+        params.entropy_thold = 2.0
+        params.logprob_thold = -0.8
+        params.no_speech_thold = 0.6
 
         whisper_reset_timings(context)
-        logger.notice("⚙️ Starting whisper transcription with VAD: \(params.vad ? "ENABLED" : "DISABLED")")
         
         if let vadModelPath = await VADModelManager.shared.getModelPath() {
-            logger.notice("🎤 VAD is ENABLED - Successfully retrieved VAD model path: \(vadModelPath)")
             params.vad = true
             params.vad_model_path = (vadModelPath as NSString).utf8String
             
             var vadParams = whisper_vad_default_params()
-            vadParams.min_speech_duration_ms = 500
-            vadParams.min_silence_duration_ms = 500
+            vadParams.threshold = 0.50
+            vadParams.min_speech_duration_ms = 250
+            vadParams.min_silence_duration_ms = 100
+            vadParams.max_speech_duration_s = Float.greatestFiniteMagnitude
+            vadParams.speech_pad_ms = 30
             vadParams.samples_overlap = 0.1
             params.vad_params = vadParams
-            
-            logger.notice("🎤 VAD configured with parameters: min_speech=500ms, min_silence=500ms, overlap=10%")
-            logger.notice("🎤 VAD will be used for voice activity detection during transcription")
         } else {
-            logger.notice("🎤 VAD is DISABLED - VAD model path not found, proceeding without VAD")
             params.vad = false
-            logger.notice("🎤 Transcription will process entire audio without voice activity detection")
         }
         
         samples.withUnsafeBufferPointer { samplesBuffer in
             if whisper_full(context, params, samplesBuffer.baseAddress, Int32(samplesBuffer.count)) != 0 {
                 self.logger.error("Failed to run whisper_full")
-            } else {
-                if params.vad {
-                    self.logger.notice("✅ Whisper transcription completed successfully with VAD processing")
-                } else {
-                    self.logger.notice("✅ Whisper transcription completed successfully without VAD")
-                }
             }
         }
         
@@ -118,19 +105,13 @@ actor WhisperContext {
         for i in 0..<whisper_full_n_segments(context) {
             transcription += String(cString: whisper_full_get_segment_text(context, i))
         }
-        // Apply hallucination filtering
         let filteredTranscription = WhisperHallucinationFilter.filter(transcription)
-
         return filteredTranscription
     }
 
     static func createContext(path: String) async throws -> WhisperContext {
-        // Create empty context first
         let whisperContext = WhisperContext()
-        
-        // Initialize the context within the actor's isolated context
         try await whisperContext.initializeModel(path: path)
-        
         return whisperContext
     }
     
@@ -138,7 +119,6 @@ actor WhisperContext {
         var params = whisper_context_default_params()
         #if targetEnvironment(simulator)
         params.use_gpu = false
-        logger.notice("🖥️ Running on simulator, using CPU")
         #endif
         
         let context = whisper_init_from_file_with_params(path, params)
@@ -160,7 +140,6 @@ actor WhisperContext {
 
     func setPrompt(_ prompt: String?) {
         self.prompt = prompt
-        logger.notice("💬 Prompt set: \(prompt ?? "none")")
     }
 }