Large Language Models (LLMs) have revolutionized Natural Language Processing (NLP) by understanding and generating human-like text through extensive training on diverse text data. These capabilities have been applied in tasks like machine translation, summarization, and dialogue systems. This research explores dialogue summarization using a fine-tuned BART LLM. Dialogue summarization presents unique challenges due to the informal nature of conversations, frequent use of colloquial language, multiple speakers, and variable contexts. In this study, the BART model is fine-tuned on both the SAMSum and DialogueSum datasets, individually and in combination, to enhance the model’s ability to handle a diverse range of dialogues. The model is evaluated using ROUGE scores, comparing machine-generated summaries to human-written ones. Results demonstrate that the model effectively captures key points and generates coherent, concise summaries. High ROUGE scores and consistent summary length across both datasets indicate successful summarization, with improved performance in handling varied dialogue styles.