Inspiration
We were inspired to create this project after realizing how inefficient and slow watching lectures are normally. As a result, we decided to make a system that is able to speed up lecture videos or other videos such as movies or films so that the user is able to have a better experience watching videos.
What it does
This app looks for silent portions in the videos to evaluate the moments in the video that require a faster speed. It is able to do this by searching every 4 frames and checks for spikes in volume.
How we built it
We built this by splitting our group into three different sectors - one for the frontend, one for the backend, and one for the final integration. Our frontend uses JavaScript, HTML, and CSS, and uses the React framework to transfer video files onto our server. The server is hosted on Google Cloud, and it detects when a video file is uploaded onto a storage bucket. It also runs a Python algorithm which uses ffmpeg and pydub to shorten the video.
Challenges we ran into
Integrating the entire system together was very challenging since we had to employ several different software to efficiently and properly integrate.
Accomplishments that we're proud of
We are proud of how fast our speed up algorithm was able to run. We were able to guarantee a speed so that the video would be able to process while sped up video was generating.
What we learned
When starting this project, we decided to all work on aspects that we were unfamiliar with so that during the hackathon, we would be able to learn a lot. As a result, we each learned about the different parts of a Full-Stack System and how everything works together to make a proper final product.
What's next for Speed Vid
We wish to continue developing this project so that it can be used in everyday environments by anyone.
Log in or sign up for Devpost to join the conversation.