Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{202780,
        author = {Dr. R. Anitha and Shivam Manoj Shukla and Sanjay Kumar S and C Sujay},
        title = {Real-Time Meeting Intelligence and Task Automation Using Edge-based AI and LLM},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {12},
        pages = {7527-7538},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=202780},
        abstract = {In today’s professional environment, virtual meetings have become an essential part of communication and teamwork. However, participants often struggle to capture important points, track decisions, and remember assigned tasks during continuous discussions. Traditional meeting assistant platforms mainly depend on cloud-based services for speech processing and analysis, which may cause delays, raise privacy concerns, and require constant internet connectivity. This paper proposes an intelligent edge-based meeting assistant system that provides live transcription, contextual analysis, and automated task handling during meetings. The system integrates real-time speech recognition, a locally deployed Large Language Model (LLM), and an agent-driven automation framework to understand ongoing conversations and respond intelligently. Based on the meeting context, the system can automatically generate summaries, extract action items, send follow-up emails, and perform information retrieval tasks. Unlike conventional cloud-centric approaches, the proposed framework performs most computations directly on edge devices. This reduces latency, enhances data security, and limits the transfer of sensitive meeting information to external servers. The modular architecture also allows easy scalability and integration with different collaboration platforms. Experimental evaluation demonstrates that the proposed system delivers accurate real-time transcription and efficient meeting assistance with minimal delay. Comparative analysis with cloud-based solutions highlights the advantages of the edge-oriented design in terms of responsiveness, privacy, and operational efficiency.},
        keywords = {Edge Computing, Agentic AI, Large Language Models, Meeting Intelligence, Real-Time Transcription, VOSK, LLaMA, Ollama, Speech Recognition, Task Automation},
        month = {May},
        }