A dissertation-sized workload
A master's or doctoral project can mean dozens of recordings collected over months. Typing them all by hand takes weeks that most students would rather spend on analysis and writing. Uploading each recording as you collect it keeps the backlog small and lets you start coding early.
Create a folder for the project, or one per study phase or site, and upload several files at once. Long recordings are split into chunks and processed in parallel. Recordings from phones (M4A), handheld recorders (WAV, MP3) and Zoom or Teams (MP4, M4A) all upload as they are.
Your conventions, applied by you
Departments and supervisors differ on how transcripts should look: participant codes, how to mark pauses and overlaps, whether to keep fillers, how to show inaudible passages, and whether to use a notation system such as Jefferson for conversation analysis. RocketScribe does not apply any of these automatically. You get a close-to-clean draft, and you apply your conventions while you review.
Write your conventions down before you start and use them in every transcript. Edit in place with the synced player, clicking a line to hear it again, or export DOCX and finish formatting in Word. Consistency across transcripts matters more to examiners than any particular style.
Checking before you analyse
Plan one careful pass per transcript. Focus on:
- Participant and place names, which you will also anonymise.
- Technical vocabulary from your field and terms participants use in their own way.
- Passages you expect to quote; confirm them word for word against the audio.
- Overlapping speech in group sessions, where words and speaker labels are most likely to be wrong.
Exports for coding and the appendix
Export DOCX or TXT to import into qualitative analysis software that accepts Word or text documents; there is no direct integration. CSV gives one row per segment with its start time, useful for your own coding sheet. Keep timestamps in your working copy so you can cite where a quote occurs and return to the recording during the viva or defence.
Speaker identification, on paid plans or with a Minute Pack, labels interviewer and participant; rename the labels to participant codes once.
Fieldwork in another language is transcribed in that language. There is no built-in translation. The Google Translate link and the AI prompt export can give you a rough reading, but translations quoted in a thesis should be your own or checked by a qualified translator.
Ethics and data handling
Check that your ethics approval and consent forms allow an online transcription service before uploading. Files are never used to train AI models, and you can delete recordings and transcripts at any time. For more on consent, anonymising and verbatim choices, see the research interview transcription page.