The rapid expansion of educational content presents significant challenges, especially for teachers and caregivers who work with specially able children. Traditional methods of summarizing educational texts are time consuming and there is a growing demand for better ways to summarize content that caters to these learners. The proposed system uses several pre-trained transformer models, like BERT, BART, T5, and Gemini Flash. Using both extractive and abstractive summarization methods, the proposed system creates summaries that are short but make sense. The performance of the system is evaluated using metrics between the machine generated summary and human summary given by the teacher. Based on the ROUGE score of precision with 0.72 and BERT score with precision of 0.90, the BART model has quality of the summaries and the relevance to the needs of students. This study shows that transformer-based summaries can significantly improve the learning experience by providing a more concise and easier understanding of educational materials.