Add voice activation (VAD) and audio fade

- Add webrtcvad dependency for real-time voice activity detection
- Create audio/fade.py with fade-in/fade-out utility
- Add VAD voice activation to client recording (sends audio only during speech)
- Apply 200ms fade-out to TTS output to avoid abrupt audio cuts
- Fix tts.py indentation error in except block

audio/fade.py

@@ -0,0 +1,34 @@
+import numpy as np
+
+
+def apply_fade(audio: np.ndarray, sample_rate: int = 24000, fade_duration_ms: int = 300) -> np.ndarray:
+    """Apply fade-in and fade-out to audio array.
+
+    Args:
+        audio: numpy array of audio samples
+        sample_rate: audio sample rate in Hz
+        fade_duration_ms: fade duration in milliseconds
+
+    Returns:
+        Audio array with fade applied
+    """
+    if len(audio) == 0:
+        return audio
+
+    fade_samples = int(sample_rate * fade_duration_ms / 1000)
+    fade_samples = min(fade_samples, len(audio) // 4)
+
+    if fade_samples <= 0:
+        return audio
+
+    # Create fade envelope
+    fade_in = np.linspace(0, 1, fade_samples)
+    fade_out = np.linspace(1, 0, fade_samples)
+
+    # Apply fade-in
+    audio[:fade_samples] *= fade_in
+
+    # Apply fade-out
+    audio[-fade_samples:] *= fade_out
+
+    return audio