diff --git a/internal/audio/c/audio.c b/internal/audio/c/audio.c
index a3fa1e62..fded47bf 100644
--- a/internal/audio/c/audio.c
+++ b/internal/audio/c/audio.c
@@ -588,7 +588,6 @@ int jetkvm_audio_capture_init() {
 __attribute__((hot)) int jetkvm_audio_read_encode(void * __restrict__ opus_buf) {
 	static short SIMD_ALIGN pcm_buffer[1920];
 	static short prev_max_sample = 0;  // Track previous frame's peak for discontinuity detection
-	static int silence_count = 0;       // Count consecutive silent frames
 	unsigned char * __restrict__ out = (unsigned char*)opus_buf;
 
 	SIMD_PREFETCH(out, 1, 3);
@@ -673,37 +672,23 @@ retry_read:
 	short max_sample = simd_find_max_abs_s16(pcm_buffer, total_samples);
 
 	if (max_sample < 50) {
-		silence_count++;
-		if (silence_count > 2) {
-			prev_max_sample = 0;  // Reset after extended silence
-		}
+		prev_max_sample = 0;  // Reset on silence
 		if (trace_logging_enabled) {
 			printf("[AUDIO_OUTPUT] jetkvm_audio_read_encode: Silence detected (max=%d), skipping frame\n", max_sample);
 		}
 		return 0;
 	}
 
-	// Detect audio discontinuity (video seek): sudden level change after silence
-	// If audio level jumps >4x after 2+ silent frames, likely a seek occurred
-	if (silence_count >= 2 && prev_max_sample > 0) {
-		int level_ratio = (max_sample > prev_max_sample) ?
-			(max_sample / (prev_max_sample + 1)) :
-			(prev_max_sample / (max_sample + 1));
-
-		if (level_ratio > 4) {
+	// Detect discontinuity (video seek): abrupt level change >5x
+	if (prev_max_sample > 0) {
+		int level_ratio = (max_sample > prev_max_sample * 5) || (prev_max_sample > max_sample * 5);
+		if (level_ratio) {
 			if (trace_logging_enabled) {
-				printf("[AUDIO_OUTPUT] Discontinuity detected (level jump %dx: %d→%d), resetting encoder\n",
-					level_ratio, prev_max_sample, max_sample);
+				printf("[AUDIO_OUTPUT] Discontinuity detected (%d→%d), resetting encoder\n", prev_max_sample, max_sample);
 			}
-			// Reset Opus encoder state to prevent mixing old/new audio context
 			opus_encoder_ctl(encoder, OPUS_RESET_STATE);
-			// Drop and reprepare ALSA to flush any buffered old audio
-			snd_pcm_drop(pcm_capture_handle);
-			snd_pcm_prepare(pcm_capture_handle);
 		}
 	}
-
-	silence_count = 0;
 	prev_max_sample = max_sample;
 
 	// Apply moderate 2.5x gain to prevent quantization noise on transients