This blog guides users through building a Voice Assistant using open-source projects such as Milvus, Llama 3.2, and various GenAI technologies including Assemby AI, DuckDuckGo, and ElevenLabs. The voice assistant is designed for voice interactions and uses an agentic Retrieval-Augmented Generation (RAG) system. Key technologies used include Milvus, a high-performance vector database, Llama 3.2, an advanced large language model, Assembly AI for speech-to-text conversion, DuckDuckGo for privacy-focused search results, and ElevenLabs for voice synthesis. The architecture of the RAG system is broken down into multiple components, each handling a specific part of the process. The system retrieves information from various sources simultaneously, including Milvus knowledge base, calendar integration, and web search fallback. The results showcase a modular design with full control, privacy-focused data management, and true ownership and control of the AI stack.