mrsean2d2 / mouthwords Goto Github PK

Currently mouthwords doesn't know how to order clips if the input has duplicate words. This will obviously need to be fixed.

I've been trying to use this but I don't get any output video. I also don't get any errors. If I understand correctly, I create a text file that has something like "I am an idiot" and have it search through video files and cut those pieces out and stitch them together. I must be reading it wrong or something.

Add an option to write a transcript to json without searching.

mouthwords should check if videos are all the same frame rate so that the user can easily avoid weird pictures.