Mike Szymaniak

To synchronize with Portuguese audio you need also Portuguese speech recognition model speech-por.zip which is missing. Assets are stored in: - Windows: %ALLUSERSPROFILE%\subsync\assets (C:\ProgramData\subsync\assets) - Linux: ~/.config/subsync/assets - Mac: ~/Library/Preferences/subsync/assets...

Added entry to FAQ https://github.com/sc0ty/subsync/wiki/FAQ#where-is-config-stored-on-disk

I was thinking about this some time ago. There are some other platforms supporting this feature (GTK3), so ideally I would use some portable library/wrapper for that.

Probably installer should link executable somewhere in the path. Will look into this. If you are using brew, you could `brew cask install subsync` and it will be available in...

I will look into this, but first I will need to prepare some setup with Fedora with Wayland.

Speech recognition is currently using CPU most and it's done by external library (Pocketsphinx) which does not support GPU processing. Other operations probably could be rewritten to OpenCL or something...

Unfortunately subsync works with very imperfect speech recognition. Only about 5-10% of words are generated correctly. Line granularity is out of reach for this approach. I'm thinking about some sort...

Snap packages are temporarily cancelled. There are some changes with snapcraft that breaks my builds.

Thank you for your work, but regretfully I have to inform you I will no longer maintain this project. See #197 for more info. Ofc you're free to fork it...

What do you mean by deprecate whole repo? I just did that