"Thank you" hallucinated

I feel like the transcriptions are getting better, but almost every transcription I do, the model adds ‘Thank you’ at the end. I know this is just how the models are trained, etc. but it’s a pain to have to delete that every time. If there is a way to account for this in the software so it can detect that remove it in a way that doesn’t remove actual transcribed content, that’d be helpful. Thanks for all your work on this!