ggerganov / whisper.cpp

Port of OpenAI's Whisper model in C/C++
MIT License
33.02k stars 3.3k forks source link

Karaoke output video has character encoding issues #791

Open DaniruKun opened 1 year ago

DaniruKun commented 1 year ago

If you try to generate a karaoke video from a source in Japanese, it will result in all non-ASCII characters being missing. Example:

https://user-images.githubusercontent.com/5202322/232864374-c6710ae8-8323-41ce-b600-db90740dda45.mp4

tnraro commented 1 year ago

It looks like the font doesn't contain Japanese.

Did you set the Japanese monospace font with the -fp option?