Planned Features

[jonathanfox5]

High priority

• Video tutorial (remember to add subs)
• Test if all json dictionaries work
• Double check that CUDA check behaves correctly on all machines

Medium Priority

• Port interactive transcript from https://github.com/jonathanfox5/translate_horizon_zero_dawn
• Frequency analysis of text documents, ebooks, etc.
• Multiprocessing in all of the places
• Think about dealing with cognates
• Is there an automated way of getting an exclude list for every language?
• Add debug function
• Notify the user if largev3 is substantially better than turbo
• Swap to the better supported python-ffmpeg
• Switch transcriber backend away from WhisperX

Low priority

• Batch convert files
• Change transcribe so that it defaults -o to -i if -o isn't specified
• Add extra information for the user if word audio files aren't available. e.g. are they in the wrong format?
• Add instructions for going from Anki to an exclude list. Probably needs a video tutorial to show the whole process.
• Add instructions for going from Lute to an exclude list. Need to do more research into how to get data out of Lute.
• Add instructions for going from LinQ to an exclude list. I've never used LinQ, would need help from an experienced user.