How to Transcribe an Interview for Your Thesis

Interviews are one of the most common data sources in a thesis or dissertation. Before you can analyze them, you need a written transcript that is accurate, consistent and easy to work with. This guide explains how to plan your transcription, which style to choose and how to move from a recording to a finished transcript.

Set your transcription rules before the first interview

Many students start transcribing right after their first interview and decide on the details along the way. This often leads to inconsistent transcripts that need to be reworked later. It is better to write down a short set of rules before you begin. Your rules should answer questions such as these:

Will you include filler words such as "um" and "uh"?
How will you mark pauses, laughter or interruptions?
How will you label speakers, for example "I" for interviewer and "R1" for the first respondent?
How will you mark passages you cannot understand, for example [inaudible 00:12:45]?
Will you add timestamps, and if so, how often?
How will you handle names and other identifying details?

Write these rules into a one-page document and keep it next to you while you work. If your department or advisor provides a template, use that instead. Your methods chapter should also briefly describe the rules you followed. See our guide on describing transcription in your methods section.

Verbatim or clean verbatim: what advisors usually expect

There are two main styles of transcription.

Verbatim transcripts record everything that was said, including filler words, repetitions, false starts and often pauses. This style is common when the way people speak matters for your analysis, for example in conversation analysis or discourse analysis.

Clean verbatim transcripts keep the full content but leave out filler words and fix obvious slips. The meaning stays the same. This style is common in thematic analysis and in many applied fields, where the focus is on what participants said rather than how they said it.

Expectations differ between fields and institutions. Ask your advisor which style is expected before you start, and note the answer in your rules document.

Step by step: from recording to finished transcript

Record with good audio quality. Use an external microphone if possible, choose a quiet room and test the setup before the interview. Clear audio saves a great deal of time later.
Back up and name your files. Store each recording with a consistent file name, such as "Interview_03_2026-10-12". Follow your institution's data management requirements.
Create a first draft. You can type the transcript by hand or use automatic transcription software to produce a first draft. Automatic transcripts save typing time, but they are not error-free and must always be checked.
Check the draft against the recording. Listen to the full recording while reading along. Correct misheard words, technical terms, names and speaker labels.
Apply your rules. Make sure filler words, pauses and unclear passages are handled exactly as your rules describe.
Anonymize where required. Replace names, places, employers and other identifying details with consistent placeholders, such as [Company A] or [City 1]. Follow your consent forms and the requirements of your Institutional Review Board (IRB) or ethics committee. This guide is not legal advice; if you are unsure, ask your IRB or your advisor.
Export and store the final version. Save the transcript in a format that works with your analysis software, for example DOCX or TXT, and keep a dated copy.

Common mistakes

Changing the rules halfway through. This makes transcripts inconsistent and harder to compare. If you must change a rule, update all earlier transcripts as well.
Skipping the review of automatic transcripts. Software can mishear words, especially names, technical terms and overlapping speech. Always listen to the recording again.
Unclear speaker labels. In interviews with more than one participant, mislabeled speakers can change the meaning of a passage.
Forgetting anonymization. Identifying details are easy to miss, especially in longer answers. Check each transcript before you share it or include excerpts in your thesis.
Underestimating the time needed. Transcription takes longer than most students expect. Plan enough time in your schedule, including time for review.

How Stillscript can help

Stillscript creates an automatic first draft of your interview transcript, which you can then review and correct. Your recordings and transcripts stay on your device. You can also export subtitles in SRT and VTT format if you want to use a recording in a presentation.

Download the free checklist