In the present digital era, meetings have become an integral part of organizational communication, collaboration, and strategic decision-making processes. Organizations across corporate, educational, and governmental sectors conduct frequent meetings to discuss ideas, monitor progress, allocate responsibilities, and make critical decisions. However, documenting meetings manually is a time-consuming and error-prone process, often resulting in incomplete, inconsistent, or inaccurate records. Such limitations reduce the effectiveness of meetings and negatively impact productivity and accountability. Recent advancements in Artificial Intelligence (AI) have enabled the automation of complex tasks involving speech and language understanding. Technologies such as Speech Recognition, Natural Language Processing (NLP), and Machine Learning (ML) have made it possible to automatically transcribe spoken conversations and analyze large volumes of unstructured data. These technologies provide an opportunity to transform traditional meeting documentation practices into intelligent, automated processes. This paper presents an AI-based meeting summary and action items generation system designed to automatically convert meeting audio into structured textual summaries and clearly defined action items. The proposed system captures meeting audio, performs speech-to-text conversion, analyzes the transcript using NLP techniques, and generates concise summaries highlighting key discussion points and decisions. In addition, the system automatically extracts action items, identifies responsible individuals, and associates relevant deadlines, thereby improving accountability and follow-up. The proposed system adopts a modular and scalable architecture, allowing seamless integration with existing digital platforms and supporting both physical and virtual meetings. By reducing manual effort and minimizing human errors. The paper discusses the system architecture, working methodology, technologies employed, advantages, applications,