Whisper stutters

Using

import mlx_whisper

speech_file = "stereo.mp3"

text = mlx_whisper.transcribe(speech_file,path_or_hf_repo=f"mlx-community/whisper-large-v3-mlx",verbose=False)["text"]

f=open("result.txt","w+")

f.write(text)

f.close()

I find that the output contains repeated phrases from time to time, enough to ruin the transcription. Eg:

... you used the address you put it in. Yes.        Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes.        Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. One day ... he arrived at the station and      he was in the village. And he was in the village. And he was in the village. And he was in the             village. And he was in the village. And he was in the village. And he was in the village. And he was       in the village. And he was in the village. And he was in the village. And he was in the village. And       he was in the village. And he was in the village. And he was in the village. And he was in the             village. And he was in the village. And he was in the village. And he was in the village. And he was       in the village. And he was in the village. And he was in the village. And he was in the village. And       he was in the village. And he was in the village. And he was in the village. And he was in the             village. And he was in the village. And he was in the village. And he was in the village. or horses        or whatever.

Maybe this is a feature of the underlying model?

Reproduce the report using the Python snippet, the stereo.mp3 input, and the mlx-community/whisper-large-v3-mlx model. Start by checking whether mlx_whisper.transcribe consistently produces repeated phrases, then compare the behavior with the underlying model or another input. Done means identifying whether this is an mlx_whisper issue or expected model behavior, with a confirmed fix or documented diagnosis.