How can I use model.transcribe for multiple concurrent request

[Guneetsinghtuli]

I am running a flask server with whisper

def detect_audio(audio,filepath):
  try:

      japanese = model.transcribe(filepath)
      return japanese
  except Exception as e: 
      raise Exception('could not detect the audio :',e)

on running multiple concurrent request to the server I am getting an exception

The size of tensor a must match the size of tensor at non-singleton dimension

It works fine for a single request but not with concurrent multiple request.

[mattcwebster]

Im look for an answer on this same topic