Voice Mode provides natural, human-like voice interactions for AI assistants, primarily integrating with LLMs via the Model Context Protocol (MCP). It supports real-time speech-to-text (STT) and text-to-speech (TTS) functionalities, allowing users to engage in dynamic conversations with their AI, whether through a local microphone or a LiveKit-based communication channel. It offers flexible deployment by being OpenAI API-compatible, supporting both cloud and local speech services.