This research proposes an integrated approach to text summarization, extending the T5 transformer model’s capabilities to include not only textual data but also audio and images. Leveraging the T5 transformer’s versatility in natural language processing, we fine-tune the model on a substantial corpus of text documents for effective text summarization. Additionally, we integrate audio and image processing modules into the system, employing speech-to-text conversion for audio and feature extraction techniques for images. Our unified summarization framework seamlessly handles multi-modal data, allowing users to input text, audio, or images, and generates cohesive and informative summaries. This innovative approach enhances the T5 transformer’s capabilities and addresses the growing demand for summarization in diverse data formats.