Transcribing gets weird at the end | Tranlate function is ENG only?

[aquaxysgit]

1. I coverted 2 hour long japanese movie to mp3 file
2. I launched app with model : 'ggml-medium.bin', GPU
3. I set output and 'transcribe'
4. It works well but in the middle of processing all dialog in movie transcribed as
   'おはようなんです'(good morning). it appeared more than thousand times in my case.
   <-- I changed model small, medium, large and not resolved
5. I selected Language: Chinese ∨Tranlate
   all dialog in movie transcribed as '(Speaking Japanese)'
   Is there way to transribing and then translating all at once?

[emcodem]

I fear there is no such mode, i would probably solve by transcribe using the best mode whisper can give and after that, translate the text using another service which can do good translation.

[emcodem]

#26 (comment)