What it does
Our web application takes a link to an article, scrapes the text, and puts it in the TikTok Subway Surfers-style video. This format is highly engaging which enables users to pay attention to the content and absorb information effectively.
Accomplishments that we're proud of
Utilizing complex libraries such as Vosk API for forced alignment of video/audio files, writing code that scrapes text from any website using the goose3 library, queried TikTok API to generate audio files, using OpenCV to overlay text based on synced timestamps, successful connection between all back-end files and the front-end, implementing support for HTTP range requests to allow for large video upload, converting video encoding to a browser compatible format
Inspiration
Maintaining focus and attention while reading educational articles can be challenging. We recognized this issue and aimed to provide a solution that makes learning more engaging.
How we built it
We used Python, HTML/CSS, Flask, OpenCV, goose3, Vosk API, and various other python libraries.
Challenges we ran into
(1) Syncing audio with individual words through extracting specific timestamps (this was very hard) (2) Editing videos using OpenCV to paste the words on a template video (3) Connecting the frontend and backend (4) Chunking the video upload from the backend to frontend
What we learned
We learned how to develop software in a team-setting and obtained a greater knowledge in the technologies that we used.
What's next for Text to Tracks
We will use our program to create videos and share them on TikTok and YouTube.
Log in or sign up for Devpost to join the conversation.