You know, there are some places that have adopted OpenAI's Whisper model and really sought to fine tune it and then keep the original audio, say, of what the doctor discussed with the patient so ...