How to Transcribe an Interview for Research, Journalism, or a Thesis

By the Speakmi team · Updated October 2, 2026 · 2 min read

Interview transcripts are the raw material for theses, news stories, user research, and oral histories. A careful workflow saves hours of editing and protects the people who agreed to talk to you.

Before you record

Choose your transcript style

Verbatim keeps every word, including "um", false starts, and laughter. It suits linguistic analysis and conversation research. Clean verbatim removes fillers and repeated words while keeping the meaning, which is better for quotes in articles. Decide which style your project needs before you start, because AI output is usually closer to clean verbatim and may need edits for strict verbatim work.

Transcribe and review

  1. Split a long interview into parts under 20 minutes so each one stays within the tool's limits.
  2. Run each part through the transcriber with the correct language selected.
  3. Turn on timestamps so you can jump back to the audio and check any unclear passage.
  4. Listen while you read. Correct names, terms, and numbers first, then fix punctuation.
  5. Label speakers by adding "Interviewer:" and "Participant:" at each change of voice. Basic transcribers do not separate speakers automatically.

Protect your participants

Replace names, employers, and locations with codes if the participant asked for anonymity, and store the original audio separately from the transcript. Using a transcriber that runs entirely in your browser means the recording never passes through a third-party server, which can simplify ethics approval and data protection requirements. Always follow your institution's own rules.

Quote carefully

Check every quotation against the audio before publishing. A transcript is a working document, and the recording remains the source of truth.

Related guides