How to Transcribe an Interview: The Ultimate Guide to Creating an Interview Transcript
So, you need to transcribe an interview. Where do you start?
To begin with, you need to know what you’re getting into. Transcribing audio isn’t simple or fast, especially if there are multiple speakers or poor audio quality, and you require accuracy and detail. Give yourself time and be patient. Ideally, start your planning before making the recording.
To get a good end result, you need a system, the right tools and an understanding of what kinds of outcomes are possible. This guide will help you build that process, and provide tips and tricks that will turn you into a transcribing master. It just takes a little preparation.
Let’s get started!
Before you start transcribing
Step 1: Understand what will affect the transcription process
First, you need to understand everything that will affect the transcription process in order for this to go smoothly. This will help you know exactly what problems you could encounter whilst transcribing the audio and how to avoid them in advance.
The speed at which you can type will affect how long it will take you to transcribe the interview, especially if the recording is long. For skilled transcribers, it takes about one hour to transcribe 15 minutes of recorded audio accurately. You need to consider how long it might take you and put aside the right amount of time for yourself, particularly if you have not transcribed anything before.
If the audio is poor quality, this will make it harder for you to transcribe, meaning it will take more time. The best way to avoid this issue is to ensure that the audio you record is high quality.
Getting a quality recording can make the transcription process much easier for you. Investing in a voice recorder is one of the best ways you can do this. However, if this isn’t possible, simple things such as ensuring you are recording in a quiet room, everyone on the recording is speaking clearly and people don’t talk over each other will also help.
The more people in the recording, the longer it will take to identify and switch between each speaker when transcribing the audio. This is another obstacle that you could face which could make the transcription process take even longer.
Understanding what may slow down your transcription process before you even record your audio can really help you once you start transcribing!
Step 2: Identify what you want out of your transcript
A big choice impacting how you will transcribe your interview is the purpose of your transcript. Is the transcript simply for reference notes? Will anyone else read it? What accuracy requirements do you have?
For example, if you simply need a transcript to sift through and pull out quotes, there is no need to capture the amount of detail required to produce a verbatim transcript that someone else could read as a substitute for listening to the recording.
Ask yourself the following questions:
- Will anyone other than me be reading this transcript?
- Can this transcript act as reference notes, or does it need to be a stand-alone document?
- Will I be able to listen to the recording again later?
- Do I already know the main points I need to capture in this recording?
- Does verbatim accuracy matter?
If you goal is to simply get a few quotes for an article, you already have an idea of what those points are likely to be, and keeping a written record of the recording is un-important, just get started. Using a dedicated transcription software platform is still advisable (detailed in the next section). However, many of the steps we will layout in this guide are designed to help people who need far more accuracy than decontextualized scribbles and a few time stamps. If you need a proper transcript, you will need more prep.
Step 3: Make sure you have the right tools
Once you understand what you want out of your transcript, you need to make sure that you have the right tools for the job.
The basic requirements are a computer with a word processor and audio player. Headphones are far preferable to speakers. You will quickly find that the isolation of headphones improves your ability to pick up details, and headphones will help cancel out the clatter of your typing. Lastly, you will want to use transcription software. This might be considered the first ‘optional’ requirement. However, even free transcription software platforms are a significant help.
You computer: You don’t need a brilliant computer to write transcripts, but it will help if you have a solid keyboard. Sitting at a proper desk will help you concentrate, and a quality keyboard will increase your typing speed. Ultimately, you should do what you find most comfortable. Just be aware that you will be typing for at least three-four times the length of your recording.
Your headphones: Good quality headphones are important. You will need to wear them for an extended period of time, and improved sound quality will help you pick out little details. Although certainly optional, over the ear headphones with noise isolating qualities are considered optimal by professional transcriptionists.
You should also consider the lead length of your headphones, making sure it is easy for you to move around. For poor audio recording, plugging in your headphones directly will slightly improve the sound quality when compared to Bluetooth.
Your transcription software: Transcription software allows you to type and control the recording within the same interface. This might seem like a small benefit, but when you think about how time consuming fiddling with your mouse would be every time you need to pause the recording, the benefits are clear. If you don’t believe me, give it a shot.
Luckily, there are several quality platforms you can access for free. oTranscribe provide this capability through an online service. Application like Express Scribe can be downloaded for free. For more features, you can upgrade to pro versions like f4 or InqScribe. These deliver integrations with additional media controls (like food pedals), different export options, voice controls and more. However, they can be expensive.
If your interest in writing transcripts is more than a one-time job, investing in better software, and even features like a food pedal are worthwhile. This will make it even easier to pause, rewind, fast forward and input timestamps.
Lastly, make sure you know how to type. This might seem like a given, but if you are a slow typist, writing your own transcript will be a painful process. You don’t necessarily need to be a flawless touch typist, but you need to be good enough. Starting out as a professional, you will be expected to have a base rate of 70 words per minute. You can get by with less, but it will make you slower.
Step 4: Think about the level of detail that will best serve your transcription needs
If you purchase a professional transcription, the first question you will have to answer is about the level of detail you require. Naming conventions within the industry are not standard. However, transcription broadly fall into three categories — full-verbatim, verbatim and other.
Full verbatim: These transcripts capture every sound. That includes false starts, repetitions, ums, ahs, interruptions and more. They might even include annotations on pauses, tone and laughter. This kind of detail is a lot harder and more time-consuming to capture. Under most circumstances, it is overkill, but if it is necessary, make sure you capture that information from the start.
Verbatim: Often called ‘intelligent verbatim’ or ‘clean verbatim’, these transcripts are edited for readability. The false starts and ‘ums’ captured in a full-verbatim transcript are omitted. This is often considered the industry standard, and is most likely the kind of transcript you will get from a transcription service firm if you don’t otherwise specify. Likely, it is this level of detail you will want to emulate when producing your own transcript.
Other: This category includes a couple different transcripts of varying levels of detail that are not considered ‘verbatim’. One of the more useful categories is ‘detailed notes’. These summarise questions and remove off-topic chit-chat, taking the editing of intelligent verbatim one step further, but still delivering a comprehensive transcript. You can also purchase summary transcripts that simply seek to summarise the main points, or draft transcripts that are basically rough drafts of final transcripts.
Knowing these different options provided within the industry can give you a number of goal posts to aim for. If you intimately know the recording, and the reason for which you will be using your transcript, you can adopt your level of detail depending on the subject matter. Think about these things beforehand so that you can direct your efforts more efficiently.
Step 5: Think about timestamps and speaker identification
Timestamps and speaker identification are two of the major additional features you need to think about when creating a transcript. The ability to know who is talking and to easily reference the recording can add a lot of value. However, this creates additional steps that will increase the time and difficulty of creating a transcript.
Again, this is something that you need to think about from the beginning. Best practice is to identify speakers wherever they change. However, if this is not something you need, avoid it. Particularly on low-quality recordings, it can be very hard to keep speakers straight but most of the time, it is important.
Timestamps are much easier. Most transcription software allows you to automatically input a timestamp with a hotkey. Some premium software simply does it automatically. Often, the best thing to do with timestamps is to try and put them in at a regular delayed interval (every 2-5 minutes) and then again whenever something important is said. Particularly if you are writing the transcript for yourself, think about the things that you might want to go back and clarify in greater detail later.
Step 6: It is time to get started
If details really matter, you need to start by just listening to the recording. If not, you can just get started transcribing. Either way, once you start writing, it is probably best to think of what you are doing as a rough draft.
Starting the transcription process
1. Listen to the recording all the way through
Now that you’re ready to get started, begin by listening to the audio recording all the way through, so you can understand its complexity and become more familiar with it. Think about the exact length of the recording and how long it will take you to transcribe it.
Listen to the recording while taking brief notes, just so you have a kind of structure for what the transcription process will look like. Identify how many speakers there are in total and consider the rate of speech, as well as if there is any technical terminology that you may not understand. Note down any other languages or accents. All of these factors might slow you down when you’re transcribing it so it’s important to keep them in mind.
Listening to the recording will also help you decide whether you want to transcribe the full interview or just parts of it. This is also a good time to decide what kind of transcript you want, as discussed earlier in the article — full-verbatim, verbatim or something less detailed.
It’s important to note that you don’t have to listen to the whole thing through, but many transcriptionists claim it can help for beginners.
2. Transcribe a rough draft
Transcribe a first rough draft. If you take a loose approach to writing your first draft, it will allow you to type a lot faster. This is important, particularly if you aren’t a great typist. You can get most of it down without worrying about being accurate and then go back to clean things up. If a lot of detail isn’t required, you might be able to just use your rough draft as the bulk of your transcript.
Try not to transcribe the whole recording in one shot — give yourself breaks to make sure you stay focused. If you’re typing the transcript up into a word processor, start the recording and transcribe as much as you can. You can pause it if you need to, but try not to rewind the recording. Going back will upset your flow of typing and slow you down. Remember, this is only a rough draft — it doesn’t need to be perfect.
One way to make the process easier is to use shortcuts. For example, many word processors, such as Microsoft Word, come with auto-correct built in. This makes typing roughly much easier — any typos will be automatically changed into the correct word.
For phrases that appear regularly, you can add set them to auto-correct and they will automatically change, saving you time. For instance, setting “bc” to convert to “because”, so whenever you type “bc”, it will automatically change, saving you time.
Adding placeholder text can also save you a lot of time when editing and formatting your transcript. You can add placeholder text for speaker names and speech habits, which can then be replaced with the full spelling using the ‘find and replace’ function. An example of this is to use the initials of a speaker’s name, or “S1” and “S2” for speaker 1 and speaker 2, which can later be changed.
As previously discussed, another way to save time is to add timestamps as you’re going through your first draft. If you know you struggled to transcribe a particular section, pause the recording and add the timestamp — such as [hh:mm:ss] — next to the appropriate parts so you can come back to it later.
This is a rough draft, but doing all of these things can help make the editing process much easier.
3. Go through it again and edit
After you have your first draft, you need to go through it again and edit it. It should be a readable transcript, but it will most likely contain typos and have sections missing so you need to rectify this.
Listen to the recording again from the start and compare it against the transcript. Make sure everything is accurate and edit it accordingly. This may take a couple of hours so it’s important to have patience, but it’ll be much easier once you already have a rough transcript made.
Using the shortcuts you set up during the rough transcript process can speed up your editing, so try to prepare as much as you can.
Once you think you have completed your first edited draft, now is the time to proofread. Replay the recording and check for accuracy. Edit these last few bits as you listen to the recording.
4. Format the document
You need to now format the transcript to look like an interview transcript. This includes adding paragraphs, headers, titles, page numbers and adjusting the font size.
Now is the time to find and replace any placeholder text that you included. Give the transcript a last quick check through and make sure it’s formatted how you want it.
Here is what an intelligent verbatim interview transcript could look like:
This is, however, just an example and you can format your transcript in the way that suits your transcript needs.
Now you’re ready to start transcribing!
We hope this guide helped you start on the path to doing your own transcription! Although the process can be tedious and time-consuming, it is possible for you to create an interview transcript.
Remember, when transcribing, try not to do it all at once. Give yourself regular breaks or it will take longer as you’ll be struggling to concentrate. Don’t feel disheartened if you think it’s taking longer than you anticipated — it’s all about patience.
However, many people do underestimate the giant task of transcription so if you feel that the transcription process is too tricky and takes too long, there are alternatives. You can hire professional transcription services to create your transcripts accurately and quickly. Many even have turnaround times of 24 hours! Leaving it to the professionals will help you save time and energy, and you can feel secure in the knowledge that you’ll get a high-quality and accurate transcript.