CSE498, Collaborative Design, Fall 2020
Computer Science and Engineering
Michigan State University

TechSmith provides software that empowers people to communicate more effectively by easily creating visual content such as images and video. Their flagship products, Snagit and Camtasia, are used by more than 30 million people worldwide.

The way people learn has drastically changed over the past year, with more and more people relying on video content for education. With this shift to online learning, a few glaring issues with video teaching arise, namely that long videos are hard to navigate and find relevant information.

Our TechSmith Video Summarizer solves this issue by automatically segmenting and summarizing any video into small, easy-to-navigate segments that the user can search through and view based on what they are seeking.

The TechSmith Video Summarizer uses speech-to-text techniques to extract the words spoken in a video into a transcript. We then use natural language processing (NLP) on this transcript to extract the main ideas of a video, and to find segments of a video all related to the same topic.

Our Video Summarizer then segments a video into a number of “mini” videos based on our NLP. Each of these segments has its own summary and keywords. Users can effortlessly determine if the content in a given segment is what they are seeking.

Allowing users to browse keywords and summaries of video segments gives them complete control over their learning and video watching experience. Users are no longer burdened with searching through hours of video to find the particular content of interest.

The front end of our web application uses HTML, CSS and React while the back end uses C# and .NET Core framework. The web application and SQL database are both hosted on Microsoft Azure. FFmpeg is used to render and break down the video.