Generating Textual Video Summaries using Modified Bi-Modal Transformer and Whisper Model

Online video material availability has grown significantly over the past few years, making it challenging for viewers to consume and process information effectively. Video summaries written in text form have shown to be a helpful technique for solving this problem. The key aspects of a video can be accurately and briefly summarised by utilizing methods like video summarising and natural language processing. The purpose of the model is to display a short textual summary of the video to the user before the user watches the complete video. It is frequently difficult and time-consuming for students to locate a suitable educational video. It is occasionally seen on video streaming platforms such as YouTube that the title and the content of the video are not precisely the same. So the model will generate a textual summary of the video and the user can read the summary and get to know about the content of the video. Also, this approach to summarizing videos is helpful in other tasks like video content moderation and could also enable visually-impaired people to consume video material and improve their quality of life.