AIMIS – AI Meeting Intelligence System

This research presents AIMIS (AI Meeting Intelligence System), a scalable and intelligent web-based platform designed to address the core challenges faced by participants in online meeting environments: difficulty responding quickly under pressure, lack of communication clarity, absence of real-time AI assistance, and the burden of manual meeting documentation. The AIMIS platform bridges the gap between AI capability and practical meeting use by integrating three-mode response generation, text-to-speech synthesis, virtual microphone routing, automated audio transcription, and Minutes of Meeting generation into a single unified interface. The cost prediction and communication assistance modules empower users to engage more effectively in meetings while eliminating repetitive manual documentation tasks. From a technical perspective, the platform is built using Streamlit for the frontend, Google Gemini 2.5 Flash as the AI engine, gTTS for speech synthesis, the Gemini Files API for transcription, and Streamlit session_state for conversation management. This stack ensures a lightweight yet powerful system deployable with minimal configuration. Experimental results confirm that AIMIS performs effectively across all primary functions, including fast and accurate response generation, reliable transcription, automated summarization, and consistent audio output. User feedback highlights practical value in reducing cognitive load and streamlining meeting workflows. This research contributes to the field of AI-augmented collaboration by demonstrating the feasibility of integrating large language models, speech processing, and automated documentation into a unified meeting assistance platform, combining natural language processing, speech synthesis, and web application development to address a complex real-world productivity challenge. Future improvements may include: •      Real-time voice input for live meeting transcription •      Neural TTS models for higher-quality, more natural speech output •      Mobile application development for broader accessibility •      Multi-user collaborative session support •      Cloud deployment and enterprise meeting platform integration

[1] A. Vaswani et al., "Attention is all you need," Advances in Neural Information Processing Systems, 2017. [2] T. B. Brown et al., "Language models are few-shot learners," Advances in Neural Information Processing Systems, 2020. [3] A. Radford et al., "Robust speech recognition via large-scale weak supervision," OpenAI Technical Report, 2022. [4] Y. Zhang et al., "Automated meeting summarization using transformer-based models," IEEE Trans. Audio, Speech, Lang. Process., 2021. [5] J. Chen and W. B. Dolan, "Conversational AI assistants in workplace communication," ACM CHI Conference, 2020. [6] X. Wang et al., "Text-to-speech synthesis for assistive communication," Int. J. Speech Technology, 2019. [7] R. Patel and A. Kumar, "Virtual audio routing for remote communication systems," J. Network Comput. Appl., 2020.