Reverted the sentence-batching SpeechQueue that caused duplicate playback. TTS now uses the original simple approach: speak full text once on Final. Retained: parenthetical content stripping + emoji removal for cleaner TTS output. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>