I really need the transcript to distinguish different voices/speakers. It is often reporting things I said as the professional, as things the parent in my session reported. It often reports the child said things that I actually said. I manually correct these afterward but if it could recognize each speaker's content separately, things would be more accurate