How to Transcribe an Interview
Transcribing an interview can be extremely time consuming if done wrong. By following some key steps and using the right tools, you can make the process easier and faster.
Before beginning to transcribe an interview, analyze the recording to understand the amount of time it will take. This will depend on the audio quality, the number of speakers, subject of the recording, and transcription style. Also account for any additional time coding and formatting required, as that will take extra time.
A simple recording (general content, 1-2 speakers, clear audio) can take 3-4 hours.
A complex recording (tech content, 3-5 speakers, clear audio) can take 5-6 hours.
2. Pick the Right Tools
After establishing the time commitment, think about the tools you will use. These should include:
Install free transcription software on your computer for audio/video playback (for e.g. Express Scribe, Transcription Buddy, or InqScribe). These programs allow you to hot-keys to easily play/pause/rewind/fast-forward the recording without having to take your hands off the keyboard to use a mouse.
To type the document, you can use the in-built text editor that your transcription software offers or use an external one like MS Word or Pages.
Be sure to learn how to save the document while transcribing because you don’t want to lose all your hard work in the event of software crashing or battery failure!
Get a good pair of headphones or earplugs to listen to the recording. You can use speakers too, but we don’t recommend that unless the audio is studio quality within a single speaker.
You will be able to capture words far more accurately and in greater detail if you use a headphone.
A Foot Pedal (optional)
If you plan to transcribe regularly, then a foot pedal can be a useful tool to save time. It basically allows you to play/pause/rewind/fast-forward a recording using your foot rather than hot-keys, saving you extra keystrokes while transcribing.
3. Follow these Transcription Steps
With the right tools in place, you can load your recording onto your chosen transcription software and begin.
Transcribe the first draft
In the first pass, transcribe quickly without worrying too much about getting everything right.
Also, don’t try to transcribe the full file at one shot. Transcribe 5-10 minute segments at a time till you have a draft of the full recording ready.
Add time codes
When you transcribe an interview, you will invariably come across words that you don’t understand on the first pass. Instead of re-playing to get everything right the first time, place a time code and mark the word ‘unclear’ or ‘indiscernible’ and move on.
Time codes are useful for quickly locating parts of the transcript that you want to research during proofreading.
Time codes are also important for your client (if you are transcribing an interview for someone else) to know which parts of the transcript has missing or possibly incorrect words, so that they can make corrections if needed.
You may also want to add general time codes while transcribing interviews (such as every 3 minutes or 5 minutes) to facilitate editing/proofreading later.
Add placeholders for speaker names
Another thing to do while creating the first draft is to add speaker names or at least placeholder text that you can later replace with actual names.
You can use any general placeholders such as 1: and 2: or I: and In: for interviewer and interviewee is enough.
When you have the first draft ready with time codes in place, replay the recording from the beginning and proofread the entire file. Again, do this in 5 or 10-minute segments so that you can concentrate better.
In this step, you will not just check for the accuracy of the words but also fill-in any missing fillers, false starts, ambient sounds etc. if the interview transcription is meant to be verbatim or true verbatim.
If there are still words that are unclear, leave the time codes in place so that readers know there is a word missing at that point. Sometimes you can insert a phonetic transcription of the word you hear (i.e. what the words sounds like to you), but even then, leaving a time code alongside is a good practice.
Add speaker names
After proofreading the file, add in the speaker names (simply ‘find and replace’ the placeholders you added in step 3 with actual names). If names are not available, use generic identifiers like Interviewer: and Interviewee:
Now you are ready for the final step in the process of transcribing an interview, which is formatting.
Unless you have specific requirements like formatting for NVivo or video editing, you can limit formatting to adding paragraphs, adjusting the font size, and adding headers, titles, page numbers etc.
Here is an example of a fully formatted interview transcript:
When your final interview transcript is ready, save it in PDF format for easy distribution.
Transcribing interviews is a very interesting exercise because of the insights it offers into the interview. You have to listen to each word that is said, which can often lead you to discover parts that you may have missed while actually conducting the interview.
However, if you are new to transcribing, this process will feel very tedious and it may initially take you 8-12 hours to transcribe an interview even if the file is perfectly clear. Don’t worry, it gets better with time as you become familiar with the tools and shortcuts available.
Do you have any transcription tips to share with those just starting out? Leave a comment to share with other readers!