forcing model.transcribe to use a particular language

[nairajay2k]

import whisper
model = whisper.load_model("base")
result = model.transcribe("someaudiofile.wav",decode_options={'language':'hi'})

Trying to force the transcription to Hindi (hi) but I get
TypeError: DecodingOptions.init() got an unexpected keyword argument 'decode_options'

Pl let me know what I need to do.

Thanks in advance

[nairajay2k]

Ok so I think when I use
model.transcribe("someaudiofile.wav",language='hi'}) it works for small but not base model. For base model it returns an urdu/persian text.

[ulatekh]

Did you try options = whisper.DecodingOptions(language = 'hi') followed by result = model.transcribe("someaudiofile.wav",decode_options=options)?