No transcription engine gets proper nouns right every time. Names, acronyms and domain jargon are where they slip — and those are usually the words that matter most. You can fix them in place.
On iPhone and iPad
- Open a transcribed interview.
- Tap the text of the passage you want to correct.
- The passage opens for editing. If audio was playing it pauses, and playback jumps back to the start of that passage, so you hear exactly the words you are correcting.
- Make your change and confirm.
On Mac
- Open a transcribed interview.
- Click into the passage you want to correct and type directly — the transcript is editable in place.
- Clicking a passage also moves playback to that point, so you can listen to the part you are working on.
Each passage is edited on its own. Work through a transcript one section at a time rather than treating it as a single block of text.
Listening while you fix
Because the audio seeks to the passage you selected, the loop becomes: select, listen, correct, move on. On a difficult recording that is considerably faster than scrubbing the timeline looking for the right moment.
While playback runs, the transcript follows along and the current passage stays in view.
Your earlier versions are kept
Whenever the transcript changes, Transcriber stores the previous version rather than overwriting it, with the date it was made. Editing is not destructive — and neither is re-running Transcribe with different settings. The transcript you already corrected is not lost when a new one is produced.
Speaker tags and timestamps
If speaker detection was on, each passage carries a Speaker 1, Speaker 2 tag. Those come from the transcription itself and are not edited here; to change how speakers are detected, adjust the setting and transcribe again — see transcription settings for one interview.
If timestamps get in your way while editing, hide them: Text only mode in the document settings on iPhone and iPad, or Toggle Timestamps in the Mac toolbar.