Work out audio to text duration instantly with clear inputs, formula shown and shareable results.
Manual transcription takes three to eight times the audio duration depending on clarity and speaker count. An hour of clear single-speaker audio takes three hours of work at 35 an hour, so 105 - about 1.75 per audio minute. A multi-speaker meeting doubles that, which is why speaker diarisation is priced separately.
Transcription time
work minutes = audio minutes x difficulty factor
Cost
cost = work hours x hourly rate
Yes - machine output plus human correction runs at about 1.5 to 2 times audio for clear speech, though poor audio can be slower to fix than to type.
Overlapping speech, cross-talk and speaker identification all require repeated rewinding, which is where the time goes.