How I Make A Video Lecture

I’ve been making video lectures for flipped-classroom music tech courses since 2016. My process for making these and also the quality of the lectures I produce has evolved a lot since then. Since many educators are suddenly finding they need to do this, I thought I’d share the basic process that I use now to make lectures.

This is just one of many great ways to make a video lecture. It is a good way, not the best way. It’s tailored to my particular aims and skillsets, which may include a bit of audio and video editing capability that is not in reach for everyone, and also is suited for music tech teaching. I hope that at least some of the process is transferable and useful for teachers in other fields with other skillsets.

I aim to create video lectures that take advantage of the medium as much as possible. I used to do video lectures that were voice-over-powerpoint, but it has since occurred to me that in video you lose so much of what is valuable in a live lecture, and so using a live-lecture paradigm in a video format is never going to be an effective replacement of a live lecture. To compensate for what is lost in the shift from live to video, it helps to think about what the video format offers that can’t be done in a live lecture.

My students watch youtube channels which are not necessarily professionally produced, but they are still fast-paced, well-edited, and very engaging, and this is what they expecting from other online video content. What I’ve found is that planning a traditional lecture and making it into a video isn’t the easiest way to achieve this style. What works for me is to plan it as a video from the very beginning.

When I first started making video lectures, I planned it like a traditional live lecture. I spoke to the camera from minimal notes, and found what many other lecturers are finding right now: there is a temptation to endlessly retake to find the right way to say something. One lecture can take days if you’re not disciplined about it. On one hand you can be firm and just commit to the first way it comes out - just as you would have to do in a live lecture. But on the other hand, you can also consider the origins of this temptation, which lie in the fact that live lectures and videos are fundamentally different formats, with different possibilities, expectations and requirements.

The process I’ve described below is more efficient than what I’d been doing previously. In reality for me, this hasn’t changed how long each lecture takes me to make, but the quality of the lectures has significantly improved. So theoretically if I aimed for the same quality as I’d been producing before, it would take me much less time this way. There are downsides - writing a full script does kill some of the spontaneity and dynamics of unscripted ad lib, so you do have to be careful not to end up sounding wooden and monotonous, or using overly formal language. And of course, this method is not immune from the endless perfectionism that plagues any other method, so you do have to draw a line at some point that says “This is good enough. This is finished.”

Anyway here’s how I do it. I hope it helps some folks.

1. Create a script
   I create a fully written-out script for the lecture. I’ll begin by planning it out in topics and sub-topics, maybe even bullet points (which can be useful later for text slides). Then I write out every word that I’ll say. I write it using the kind of language in which I speak, so it doesn’t sound like a read-out essay.

   This helps to achieve the right length from the beginning - a 30-minute lecture for me is usually about a 3500-4000 word script.

   I then colour-code the script according to what will be shown for each segment of the lecture. E.g. Red text for any diagrams or animations I plan to create, blue for screen capture sections, and green for where I plan to shoot my talking head.

   For me it usually ends up being about 10% talking head, just at the start, end, and for some segues and more personal moments. Then it’s about 20-30% animations, diagrams and screenshots. The remainder will be simple text slides.

2. Record the script
   Then I make an audio recording of the script read-through. This usually doesn’t take long, because I don’t have to think about how I’ll phrase things, I just read. I’ll also make any edits to this audio so that I have a single clean edit of the full lecture.

   Note that I record the talking head parts as well, even though I’m likely to have to retake those when I record actual footage of myself talking. I do this for two reasons: it helps me to plan timing, and it is insurance against me running out of time to shoot the talking head.

3. Create or source diagrams, animations, and screen captures
   Using the colour-coded script as a guide, I create any of the special visual assets that I plan to show. I use apple Motion to create simple animations to visualise concepts (though this is probably not something everyone can or should do), and I capture screen recordings using quicktime or Apple’s new “screenshot” app.
4. Create titles and text slides
   Next I create all the text assets I need. I do it in Final Cut Pro X, because that’s what I edit in, but you could just as easily do it in Keynote or Powerpoint and then export your slides as images. I make sure I save any of the bullet-point summaries that I made when writing the script, as these can often just be cut-and-pasted to make text slides.
5. Edit
   Use video editing software (I use Final Cut Pro X but you could use anything, including iMovie) to put things together on a timeline. I place the full audio track in first, and then arrange the other assets around that.
6. Record talking head segments
   Then I record my face talking. The reason I do this so late in the process is because I see it as nice but expendable. It can be time-consuming, and whether you can actually get it done can be contingent on environmental factors. If you run out of time to do this part, the lecture will still be ok. Since you already recorded the text, you can just pop some slides over it and the lecture is done. But if you do get it done, you can easily pop it into the timeline where you have saved a place for it.
7. Final edit and master
   With all the assets edited and arranged, you might need to give it one last watch-through to tidy up any glitches, and importantly, make sure the audio levels remain consistent throughout the lecture. E.g. if you’ve got some audio from filming your talking head or other audio excerpts, you need to make sure the volume of those matches the audio from your main script recording.