How to Transcribe an Interview for Free (Without Uploading It)

Transcribing an interview by hand takes about four to six hours for every hour of audio. Automatic transcription cuts that to minutes, and the result is good enough to work from. The catch is that most transcription services upload your recording to their servers, which is a real concern for confidential interviews with sources, research participants, candidates or clients.

This guide covers how to record an interview so it transcribes well, how to get a free transcript without uploading the recording, and how to check and format it so you can quote it with confidence.

Before the interview: record for the transcript

Most transcription errors come from the recording, not the software. A few habits make a big difference:

• Get the microphone close. A phone on the table between two people works far better than a laptop microphone across the room. For one-to-one interviews, a small clip-on microphone for each person is ideal.
• Choose a quiet room. Cafés, open offices and rooms with echo all make speech harder to recognize. Turn off fans and music.
• Avoid talking over each other. Overlapping speech is the hardest thing for any transcriber, human or machine. Leave a beat before you ask the next question.
• Record calls directly. For Zoom, Teams or Google Meet interviews, use the app’s own recording (with permission) instead of holding a phone to the speaker.
• Ask for consent on the recording. Start by stating the date and asking the interviewee to confirm they agree to being recorded and transcribed. It’s good practice, and required in many places.

Option 1: transcribe it in your browser, without uploading

Modern browsers can run speech recognition models on your own computer. ClipCocoon’s interview transcription works this way, and it’s free:

1. Open the tool and add the recording: MP3, M4A or WAV from a recorder or phone, or a video file such as MP4. Interviews up to an hour work in one go.
2. The first time, the tool asks before loading its speech model, which your browser keeps for next time. Then it transcribes the interview on your device.
3. Read the transcript under the player. It comes in paragraphs with timestamps, and a pause starts a new paragraph, so questions and answers usually separate.
4. Fix names with Find & replace, and click any word to correct it.
5. Download a Word document or plain text, with or without timestamps.

Because the recording never leaves your computer, there is no retention period to worry about, no terms granting a company a license to your audio, and nothing to breach. It currently transcribes English only, and it doesn’t label speakers. On a recent laptop in Chrome, Edge or Safari, an hour of audio takes roughly 8 to 15 minutes.

Option 2: an online transcription service

Services such as Otter, Rev and TurboScribe upload your file and transcribe it on their servers. Many offer speaker labels and other languages, and some have human transcription for difficult audio. Before using one for a sensitive interview, read its privacy policy for three things: how long recordings are kept, whether your content may be used to train their AI, and who else processes it. Our guide on what happens to uploaded files explains what to look for.

Option 3: transcribe it yourself

For a short interview, or one with poor audio, typing it yourself is still an option. Use a player that can slow down playback, and a keyboard shortcut to pause and rewind a few seconds. It’s slow, but you’ll know the material well afterwards. A good compromise is to start from an automatic transcript and listen through once, correcting as you go.

After transcribing: check before you quote

An automatic transcript is a draft. Before you rely on it:

• Fix names first. People’s names, company names and jargon are the most common errors. Replace each one everywhere at once.
• Listen to every quote you use. Click the paragraph’s timestamp and hear the passage again. A single misheard word can change the meaning.
• Watch for missing words in crosstalk. Where two people spoke at once, the transcript may skip one of them.
• Mark unclear passages. If you can’t make something out, write [inaudible] with the time rather than guessing.

Formatting the transcript

How you format depends on what the transcript is for. Journalists usually need accurate quotes with times. Researchers often need speaker labels and line numbers for coding, which are easy to add in Word: download the .docx, then add initials such as “Q:” and “A:” at the start of each paragraph and turn on line numbering. For publishing, an edited transcript reads better: remove filler words like “um” and “you know”, false starts and repetitions, while keeping the meaning.

If the interview is also a podcast or video, the same transcript can become subtitles: download an SRT file, or burn captions into the video with Auto Captions.

Storing interview recordings safely

Whatever tool you use, think about where the recording and transcript live afterwards. Keep them in an encrypted folder or drive, delete copies you no longer need, and follow any data rules from your newsroom, university ethics board or employer. Transcribing locally removes one copy from the picture: the one a transcription company would otherwise have kept.