Inspiration

Many students experience frustration when reviewing lengthy lecture recordings, particularly when revisiting material they have already mastered. The tedious and inefficient process of sifting through hours of lecture content, coupled with the recent popularity of short-form content, inspired us to create LockIn. We envisioned a world where dull lectures could be transformed into engaging, bite-sized pieces of content that are equally effective, to help students learn faster, smarter and better, together.

What it Does

LockIn takes lecture videos uploaded by users and analyzes them to identify and extract highlights. These highlights are the most critical parts of the lecture, such as key concepts, important explanations, and summaries. By providing a concise, curated summary of the lecture content, LockIn helps users quickly review and reinforce their understanding without watching the entire video. After the videos have been watched by the user, they can choose to engage with the Community tab, which allows them to discuss and collaborate on related units they are studying, fostering a collaborative learning environment. Users can connect with peers to share insights, ask questions, and deepen their understanding of the material with like-minded individuals that use our app. Discussions are fast and easy as students do not have to sift through hours of lecture videos to find that one segment that is being discussed on the forums.

How We Built It

We built LockIn using a combination of video processing and natural language processing (NLP) technologies. The backend is powered by a machine learning model trained to recognize important segments in lecture videos. We used whisper API to generate a JSON transcript of the extracted audio. We used Python for the server-side logic and integrated various libraries for video handling and NLP. The frontend is a user-friendly interface built with flutter, allowing users to easily upload videos and view the generated highlights.

Challenges We Ran Into

One of the main challenges was accurately identifying the key moments in the lectures. Lectures vary widely in style and content delivery, making it difficult to create a one-size-fits-all solution. Visual Studio Code does not support AAC audio encoding format, so we had to find creative solutions to bypass these technical limitations when uploading videos to our app. Finally, we also had to ensure that the application could handle large video files efficiently, providing quick and accurate results without excessive processing time.

Accomplishments That We're Proud Of

We're proud of creating a tool that can genuinely help students save time and improve their learning experience. It was particularly rewarding to see our application begin to successfully identify crucial parts of lectures, convert them into short-form content and stored to the cloud to be shared with . Another accomplishment was building a smooth, intuitive user interface that makes it easy for users to interact with the app. We are also proud of making a platform that can potentially bring students and alumni alike together to clarify and provide insights into certain parts of a lecture that may be deemed trickier to grasp by others.

What We Learned

Throughout the development process, we learned a lot about video processing, editing and trimming, audio extraction, NLP, and machine learning. We gained a deeper understanding of the challenges involved in working with multimedia content and the importance of user experience design. We also learned the value of iterative development and user feedback in refining our product.

What's Next for LockIn

Looking ahead, we plan to improve the accuracy and efficiency of our highlight extraction process. We also aim to expand the functionality of LockIn by adding features like keyword search, note-taking, and a quiz generator. Additionally, we want to explore mobile app development to make LockIn even more accessible. Our ultimate goal is to create a comprehensive study aid that supports students in their academic journeys together.

Built With

Share this project:

Updates