Speech to text with a human review pass, speaker labels and timestamps, delivered as a document plus an SRT caption file for video. Standard turnaround is 48 hours.