Abstract: In an era of information overload, automatic text summarization has emerged as a crucial tool for efficiently extracting significant insights from large volumes of text. The development and deployment of a multilingual abstractive text summarization system driven by artificial intelligence is examined in this work. There are two primary approaches to summarizing: extractive, which uses key phrases directly from the source material, and abstractive, which constructs new sentences to make the key ideas easier to understand. In order to provide more logical and concise summaries, our project uses an abstractive summarization model, which rephrases the input content rather than just selecting portions of it. The system includes features like text-to-speech conversion, automatic language detection, and translation in addition to processing documents in Kannada, Hindi, and English.This all-encompassing strategy seeks to improve usability and accessibility, especially in environments with limited resources and multiple languages. The resulting summaries show how abstractive approaches can perform better than extractive ones in terms of readability and contextual relevance.

Keywords: Automatic Text Summarization, Abstractive Summarization, Extractive Summarization, Artificial Intelligence, Multilingual Summarization, Text-to-Speech Conversion, Language Detection,Contextual Relevance, Document Processing, Natural Language Processing (NLP), Summarization Model.

| DOI: 10.17148/IARJSET.2025.125242