pajowu

Hey, sorry for not responding sooner, we all were a bit busy over christmas and new year. This is a kind-of known issue: The transcription happens over the entire file...

Hey, just checking in to see if you we're able to try this again with a recent version and speaker diarization? Is this still an issue for you?

This release is broken on linux (#277), we are planning on releasing a new prebuild with a fix this evening

The inference part is still missing from what I can tell from a quick look at the code (i.e. the part that actually uses whisper to transcribe something)

Hey, thanks for opening this issue. I'm not quite sure I understand the first one: Do you mean having multiple lines in a cue (i.e. a subtitle displayed at the...

This is a great idea. Howevery we would only be able to release them on a best-effort basis as we currently do not have the hardware to test it on

Okay, I did a quick investigation to figure out what would be needed for this. There are multiple aspects, which I describe below. As this is a larger topic, I...

Hey, just popping in real quick with some good news: We just published the first release with working builds for apple silicon. This implements "Apple Silicon" from the above list....

Hey, yes, speaker diarization really needs to be improved. The current models varies widely in its results depending on the difference of the speakers within the audio and the level...

This might be a good moment in general to think about how we store media files. At the moment they are always in ram, which is not ideal i think