With the rapid increase in the volume of text documents, identifying relevant information on the web is becoming increasingly challenging. Automatic Text Summarization (ATS) addresses these challenges by efficiently processing large volumes of documents and extracting the most pertinent information from them. Despite its significant advancements, ATS still encounters several challenges, including handling long and repetitive sentences, maintaining cohesion, and ensuring semantic similarity. The current work proposes an extractive text summarization technique utilizing topic modeling, aiming to generate summaries that feature highly representative sentences, minimal repetition, concise length, and strong semantic similarity. The effectiveness of the proposed technique is validated through experimental results on DUC datasets, showing that it outperforms conventional methods.