Meeting Summarizer Using NLP

This project proposes a hybrid Natural Language Processing (NLP) system for automating meeting summarization and action item extraction. The system converts meeting audio into text using speech-to-text technology, preprocesses the data, and applies a combination of extractive and abstractive summarization techniques utilizing transformer models such as T5 and BART. Additionally, it extracts critical action items including tasks, deadlines, and responsible individuals through named entity recognition and dependency parsing. The designed web-based platform enhances productivity by delivering clear, concise meeting summaries and structured follow-up actions. Evaluation on real and synthetic datasets demonstrates improved accuracy and effectiveness over traditional methods, making it a valuable tool for efficient organizational communication.