Word-by-word captions

[xezpeleta]

With the one-level timestamps JSON generated by WhisperX, we can generate videos with captions like the following:

https://youtu.be/kKxggfjW7d8

pip install word-by-word-captions
wbw-captions --video file.mp4 --json file.json --output file_w_captions.mp4