Video to Text conversion and Abstractive Summarization for Effective Understanding and Documentation

Abstract

Video is one in every of the sturdy sources of data and the consumption of on-line and offline videos has reached a new level within the previous few years. An elementary challenge of extracting data from videos is, a viewer should undergo the whole video to grasp the context, as against a picture wherever the viewer will extract data from one frame. Typically, protracted videos also are quite troublesome to follow because of reasons like totally different pronunciation, pace then on. Abstractive Text summarization extracts the utmost important information from a source which is a text and provides the adequate outline of an equivalent. The analysis work conferred during this paper describes a straightforward and effective methodology for video Summarization. It principally targets academic and technical videos. Speech is extracted from video. The speech is regenerate to the corresponding text using abstractive summarization technique and produces summarized text. For quicker conversion of video to text GPU can be used. This has numerous applications like lecture notes creation, summarizing catalogues for protracted documents then on.