langswap-app/langswap ? reverse-engineered prompt
Reverse engineered prompt
Build me a self hosted web app that can take a video, figure out what is being said, translate it into another language, and then dub the video with a cloned version of the original speaker’s voice.
I want a simple browser interface where I can upload a video, choose a target language, and get back a finished dubbed video plus subtitle files for both the original and translated speech. It should run locally on my own GPU, with no cloud service required, and it should work end to end in one place. If there are options for different voice models or timing adjustments so the dubbing feels more natural, include those too.
Make it easy to run with Docker, and also give me a way to run it from source for testing. If you need to check current setup details or model docs online, go ahead and do that.