This project was born out of a common but overlooked problem: many videos, especially older or fan-translated ones, only contain hardsubs—subtitles burned directly into the video. These are impossible to toggle off and difficult to extract. At the same time, many creators or archivists need to reuse or translate subtitles, but don’t have access to the original subtitle files.
We wanted to create a system that solves both parts of the problem:
Generate hardsub videos from soft subtitles at minimal cost.
Build a pipeline to extract hardsubs back into soft subtitles using modern tools like OCR and AI.
Log in or sign up for Devpost to join the conversation.