As Online classes have been on a boom due to Covid-19 Pandemic, teachers have not been able to conduct classes in a more inclusive way and lack coordination with the students. As a result, students have not been able to grasp concepts accurately of the lectures thereby lacking the required knowledge on student's part. This inspired us into making a lecture summarizer which can be leveraged to provide lecture summaries and score students based on the summary provided.

What it does

  • Initially, The teacher or the user provides a lecture recording of 3 hours maximum to the webapp.
  • Next, the application goes through the recording and extracts key highlights of the lecture as a summary
  • Subsequently, the students are asked to provide lecture summary and are scored on the basis of how they performed when compared with our application for providing a near to accurate summary of the lecture. ## How we built it
  • We have made use of HTML, CSS and JavaScript for the front-end part.
  • We have made use of Django for the backend and library called DeepSpeech to extract the voice and convert it into text format.
  • DeepSpeech is an open-source Speech-To-Text engine, using a model trained by machine learning techniques based on Baidu's Deep Speech research paper. Project DeepSpeech uses Google's TensorFlow to make the implementation easier.
  • Based on the text pointer are generated as key highlights of the lecture .

Challenges we ran into

  • We were able to complete the project without any difficulties.

What we learned

We have learnt how to develop fully fledged application as well as how to integrate backend with front-end. We have also learnt how to leverage NLP libraries for our project and provide great solutions.

What's next for Erudition

  • We would like to further modify and recreate the app to output important questions retrieved from the lectures so that it helps teachers as well as students to obtain the key points and help students in their preparation for exams.
Share this project: