April 2026 monthly summary for vllm-omni (vllm-project/vllm-omni). Focused on enabling real-time TTS streaming by adding PCM audio chunk streaming support. Implemented the _generate_pcm_chunks method in OmniOpenAIServingSpeech to support streaming PCM chunks for text-to-speech functionality, laying groundwork for low-latency, interactive voice experiences. This work is tracked under commit 2c6c07c4f68385cf4f625a5cdb0dec710e9c0fff (PR #2569) with full sign-off, demonstrating strong code discipline and collaboration. Key business/value impact: reduces end-to-end latency for TTS, enhances user experience for voice-enabled applications, and establishes a reusable streaming pattern for future PCM-based enhancements. No major bugs fixed this month; the focus was on delivering the streaming feature and preparing the foundation for future enhancements.