noScribe works extremely well if we have prerecorded audio but sometimes we have live presentations that we want to transcribe and translate on the fly, is there a chance that this will be supported in the future or is that totally out of scope?

While there are other tools like https://github.com/chidiwilliams/buzz for that and usually tools are best that have a small and focused goal, I want to argue that this might be a different case:

1. Behind both noScribe and buzz there is the same underlying Whisper model which offers this functionality.
2. Both tools have large download and install sizes, which means if one needs one and only has the other it means both a long download time and also a lot of space on the storage medium.
3. My colleagues and I find buzz quite difficult to install.