Video Learning Blog 9 – Interview Transcripts

When working with students in an in-person training, we spend some time talking about how to transcribe interview audio ourselves, by hand on our computer keyboard. The reason for this is that there is no better substitute, especially for the new filmmaker, to doing the work yourself. It will take you some time, but it will save you a great deal of time in the story building and editing process.

The gist of creating your own transcript is pretty simple – and can be made easier with a number of tools, such as a transcriptionist’s foot pedal, which allows you stop and start the audio easily to keep up with your typing ability. You can also playback your audio at a reduced rate of speed, so that it’s easier to keep up with fast talking interviews.

There are a number of online services that provide affordable, professional transcription service in quick turn around times. I often use these, especially on professional projects where there is a limit to the time I can spend on such tasks, or there are several interviews and a looming deadline. It’s not the best way of doing it in terms of quality of the editing experience, because you are ultimately more familiar with the interview when you do it yourself. But, it works.

As technology moves forward, there is also an increasing number of automatic transcription apps that can be utilized. You can find many examples of how this might work on youtube – usually it requires you to listen to the interview in headphones, and say the words the person says for the app to be able to create a text version using voice recognition tech. While tempting to try, I have often found that the acuity of the technology is not there – leaving you with lots of minor mistakes and formatting issues to go back and fix afterwards. For me, this additional work and loss of accuracy makes these apps a non-starter.

The WHY of Transcripts

Regardless of how you produce your transcript, it’s most important to focus on why you’re actually doing it. Even with only one speaker, your interview may have taken 30 minutes or even an hour. Cutting that down to 3 to 5 minutes can be a mind numbing experience, and result in a great deal of wasted time scrolling back and forth in the editing software.

While working from a transcript will speed up the time it takes to accomplish your edit, it has other benefits too. It allows people without the technical skill to participate directly in the editing software to have a say in what the story looks like. They can quickly read through the transcript, and help select individual soundbites that they like. If you are working as a producer or director of a project, where somebody else is executing the assembly of the edit – this is going to be critical. You can also share this paper edit with a client, or even the person whose story you are sharing – so they can approve the direction before you invest a great deal of time into editing.

Before I move to creating a paper edit, I’ll review my story board for the project, and go through the transcript and apply a bold font to the sections I might want to use. Once I have gone through all the elements of my story board, and highlighted all the sound bites I’d like to potentially use, I can copy them to another document, sometimes categorized under the individual scenes I laid out in the storyboard. I call this organized document the “bull pen” of potential soundbites. This is the first phase of a paper edit.

An Example

As part of our Indie America series, we decided to make a video about a college professor who is also an avid dumpster diver. Like I’ve explained in earlier posts, we shot these a day or two before an in-person training – in this case at KERA in Dallas, Texas – and went through the editing process directly with our students over the course of the training.

You can download the full transcript of the interview here. In this case, we produced this document the day of the interview, afterwards in a hotel room. The overall transcript is over 6,000 words. Without a transcript, this would be a monumental amount of video to wade through and find a story.

In addition, here is the paper edit of the story we created working from the transcript.

The Paper Edit

Once we have the “bull pen” of all our potential soundbites you want to use, we can begin to order them by subject matter, and begin structuring the paper edit. This is where the fun and creativity of the editing process really begins. But, again, it leans heavily on the work you did in the pre-production portions of the process to go smoothly.

If you take a look at the paper edit I created for our example video, you can see length here has been winnowed down to about 650 words. This is still over-long for the length of our intended video. Leaving some additional material to be cut out as we fine tune the edit is pretty common for this type of project. It’s easier to cut something than have to go back to the transcript looking for more material later.

As a general rule, a video of this length is going to be about 300 – 400 spoken words in a paper edit. In our industry, we know that a professional announcer can read about 2 words per second aloud. However, a video also has lots of ebbs and flows, musical interludes or bumpers, and other times when the speaker isn’t talking. This “breathing room” is an important part of the flow of a video, and will eat up quite a bit of runtime.

If you want to take an additional step towards a complete paper edit of your video, in italics or another font, you can assign each section of spoken dialogue a series of images that will be seen on screen during the scene. These can be taken straight from the story board, if you are confident you captured those visuals. This will help us in the edit, but also help those working with us to visualize how the final piece will look.

Opening Hooks and Closers

One thing that transcripts are not great at doing is communicating the tone of the individual soundbite. Something that looks great on paper may have been said in such a way that it doesn’t actually fit when you go do the edit. This is another reason why producing your own transcripts can be valuable. As you go through the interview, you can mark or denote sections that are said in such a way that they lend themselves to good “opening hooks” and “closers”.

An “opening hook” is the type of soundbite you would choose to open the video, or start a new scene. Oftentimes, it’s not particularly interesting to a viewer to have the speaker open a video or scene with something too “on the nose”. So, rather than have the opening lines of the edit be the person saying their name, or what they’re doing – it’s better to find some type of hook that can draw the viewer into the scene with the type of information that is not easily conveyed in the b-roll images or with text on the screen.

For an opening hook soundbite, I try to look for something that is unexpected. It may reveal more about what the heart of the video is going to be about, and give the viewer a deeper sense of who the person is. Often these opening lines will come later in the interview, or further into the transcription. If you go back and review the video and transcript examples, you’ll see that the opening line is:

“oil, gas, cattle and cars … theres a lot of red meat out there, there are a lot of 8 lane freeways, theres a lot of oil and gas money floating around in Fort Worth”

I chose this opening line because it speaks to the sense of place where the video will be, but also has a deeper way of showing who the speaker is. In Dallas/Fort Worth, with big trucks and 6 lane highways – he’s on a bike when we hear this opener. This wasn’t the first thing he said, and came about halfway through the interview… as we reached more philosophical questions.

The “Closer” is the final word that the speaker says in the video or scene. You can use this for all kinds of different effect, but the main thing is you need a statement that has a full stop. That means the auditory sentence ends in a period – a completed thought. There are elements of the tone of voice that lend themselves to a good closer. In this case:

“Don’t look at me as though i’m the one who’s figured it out, i’ve just figured out one way doing it, but theres a thousand other people doing it too.”

I selected this as the closer for the video because I felt like it reinforced the underlying premise that while the subject is a PhD professor, he is also an incredibly humble and down-to-earth character. Because of the way the line was delivered, I was able to end the video showing his face, including the brief moment after he finishes speaking. That is the kind of thing we look for in a good closer.

Internal Edits

One of the ways you can shorten a soundbite is by removing portions of it – often in the center of the bite itself. This means you cut out a few sentences within a single response to shorten it, while working to maintain the original intent of the response.

When you compare my initial paper edit to the final video in the example, you can tell that I have done a number of internal edits, to shorten the overall length and make his responses more concise. If I know I want to do this during the paper edit phase, I’ll denote the internal cut in the soundbite text with a simple “…” to indicate the removal. This doesn’t always play out the way you want it to in the edit, but it’s another tool you have at your disposal.

It is also possible to combine two separate responses into a single one, in what we might call a “franken-bite”. One needs to be careful in doing this, again to make sure that the speaker’s original intent is kept, and not taken out of context. This can be difficult to do, but it does happen. My general rule of thumb on this is if I, as the editor, can tell there has been a cut there – then it doesn’t work. Maybe the tone is off, or sounds awkward. If it flows enough to not be an obvious edit – and the speaker themself might believe it was something they said, then it’s likely OK. But, for beginners, or when dealing with sensitive subjects, it is often a good idea to avoid this.

A Quick Exercise

After watching the example video, download and spend some time with the transcript. As you scroll through, look for the soundbites that I selected. You can refer back to the provided paper edit for this, too.

Now, working from the transcript – take some time to create your own story edit. One of the fascinating things you’ll find is that you could choose to create multiple, different types of stories from a single interview. Many of the decisions I made go back to my original story boarding, and especially the shot list and scenes I was able to capture in the shoot.

Remember, you can’t really tell a story if you don’t have the visuals to match up. While working with a transcript, you realize you have many options – but all your decisions in building the story have to be supported by the visuals you have.