Blind and visually impaired individuals encounter significant challenges when accessing search results online, primarily due to the vast amount of information available and the lack of adequate support tools tailored to their needs. While existing solutions like screen readers facilitate sequential exploration of search results and some tools organize results based on user behavior, there remains a gap in providing efficient online summaries of queries specifically tailored for this demographic. This study proposes an integrated system designed to address this gap by providing summarized search results, with a focus on accommodating the Arabic language. Two distinct approaches to online query-based summarization were introduced. The first approach aims to identify the most relevant sentence to the query from the search results, while the second approach utilizes an adaptive technique to extract the most pertinent sentences from the first 'n' documents in the search results. Comparative evaluations were conducted, with the first approach being benchmarked against Google Assistant, while the second approach was assessed based on summaries provided by two human experts and GPT-4. Results indicate that the first approach generally outperformed Google Assistant. Furthermore, experimental evaluations demonstrated the close alignment of the results retrieved by the second approach with the summaries provided by GPT-4, with an average relevancy score of 0.92, and the highest similarity scores of the second proposed system with Experts 2 and 1 are 0.93 and 0.85, respectively. These findings underscore the utility of the proposed system in facilitating access to information for blind and visually impaired individuals.