Fix for loop of repeating text

[mlndig]

Hi,
I've repeatedly ran into the issue of getting stuck in a loop of repeating text, using the large-v2 model (precise).
I originally dealt with it by transcribing shorter sections and joining them manually, as suggested in the Readme.

But then I found out about the "condition_on_previous_text" parameter, which when set to false gave me pretty good results and so far I haven't encounter any of these loops when using it.
Here's what it says in the faster-whisper transcribe.py file :
condition_on_previous_text (bool): If True, the previous output of the model is provided as a prompt for the next window; disabling may make the text inconsistent across windows, but the model becomes less prone to getting stuck in a failure loop, such as repetition looping or timestamps going out of sync.

So I've added the parameter to the list of parameters of model.transcribe on line 1189, so it now reads :

segments, info = model.transcribe(
                        self.tmp_audio_file, language=whisper_lang, 
                        beam_size=1, temperature=0, word_timestamps=True, 
                        initial_prompt=self.prompt, vad_filter=True, condition_on_previous_text=False,
                        vad_parameters=dict(min_silence_duration_ms=200, 
                                            threshold=self.vad_threshold))

and that does the trick.

Perhaps it would be good to have the option added in the GUI so people can choose whether to use it or not ?