i'm sure that there's a simple way to get the API to do this but i figured i would ask here

[Leggyweggy]

what would i need to add to the code to get the transcription to auto translate to English? i have another command line tool that uses whisper that accepts audio files and a bunch of different arguments for what whisper should do with it, including translation in the form of the "--task translate" argument. in the main transcribe_demo.py file i see where a couple arguments are being set for program, so i simply tried adding a similar line containing that argument, but i couldn't get it to work.

thoughts?

[davabase]

Change this line from

result = audio_model.transcribe(temp_file, fp16=torch.cuda.is_available())

to

result = audio_model.transcribe(temp_file, fp16=torch.cuda.is_available(), task='translate')