EasyA+ with YouVersity

Our website segments lengthy videos into shorter clips (sub-topcis) based on the content of the video. Our product help students conduct more effective learning during remote learning environment.

Comment

Inspiration

The idea is to increase the accessibility of information to students at all levels. Especially, with the remote learning environment, more and more students rely on online video resources for gaining knowledge. Lengthy videos may be time-consuming and distract the students from the main topic. The idea is to segment lengthy videos into shorter clips based on the content of the video. Students only need to watch the relevant videos and can save time and get the required information. Different software, APIs, and tools are used to extract, process, and display the videos.

What it does

The main features of the project are as follows:

  • Takes video URL as an input and transcripts the audio
  • Segments the video based on different topics discussed in the video
  • Extracts keywords for each segment of the video
  • Provides URLs for the extracted topics
  • Includes a web-based UI to facilitate the users

How we built it

Following are the tools used in this project.

Challenges we ran into

A Github repository was not running on our local machine due to package version variations that couldn't be solved and took a lot of time.

Accomplishments that we're proud of

We are proud of the fact that despite the short time, we were able to complete the project and submit it in running condition. This can be helpful for students in searching for relevant information and save their time and keep them on track.

What we learned

During this hackathon, we have gained experience in back-end and front-end development. We learned how to integrate our model into the website. Most importantly, we learned about neural language processing. Such as extracting transcripts from a video file, and extracting keywords from a text file. Also, we learned how to do topic segmentation for text files.

What's next for EasyA+ with YouVersity

  • Improve the automatic topic segmentation model, e.g., apply advanced NLP models (self-attention)
  • Summarize the segmented paragraphs with assigned topics
  • Automatically generate summarized reports from the input video for selected topics
  • Group topics together for easier accessibility

Built With

Share this project:

Updates