Creating a Simple Slideshow Voiceover
In a much-needed effort to encourage and promote facility of new media among its attendees, the most recent American Association for the History of Medicine annual meeting (Chicago, 2014) held a screening of video shorts (complete with an old-timey popcorn machine). Having wanted to play around with video editing for some time, I could hardly resist this legitimization of my procrastination.
In the conversation that followed the screenings of selected videos (the standard session-sized room was packed and lined with people standing along the sides and back), it was clear that scholars hadn’t embraced the video was that it was too foreign, too technical, and too outside the boundaries of the historical profession. The technology was simply too time consuming to learn (if not use), the payoff too uncertain. Yet this was entirely contradictory to my own experience creating a scholarly video for the first time. Perhaps this brief explanation of the process could lower the barrier to entry for others.
Although not having made any kind of video before, my own hubris and ineptitude prevented me from shying away from digital challenges, and I was not intimiated by the prospect of learning new software or techniques. I was prepared to suffer. To my surprise, figuring out the content and story of my video, something I thought I could do well–having worked with the material for, say, the last decade–was actually far, far more difficult than any technical challenge. More importantly, it afforded me a fresh perspective on my own research that the medium itself enabled. Limitations do indeed lead to creativity and insight.
The intent here is to outline a general strategy for creating a narrated slide-show voice-over. It’s not the only approach to making a video version of something; it’s not the best; and in fact it might not be very good at all. The hope is to show that conceptualizing and executing the technical bits is actually quite straightforward and cannot in good conscience be used an excuse for not engaging with the technology, especially as it facilitates broader dissemination of humanities scholarship.
Keynote: Without animation skills or software or time to learn something totally new before the submission deadline, I went with what I already knew. My whole “movie,” then, is little more than a narrated slideshow.
Camtasia: Very good software for creating screencasts and editing them. I used the free trial, but I eventually bought the full version since I wanted to keep making stuff, and it was so easy to use.
Although I used Keynote, which is of course exclusively for Mac, nothing in this procedure is (to my knowledge at least) unique to either Keynote or Mac; rather, it outlines a general strategy that could be used with any combination of presentation software and operating system.
Making things move
Within Keynote (though, again, Powerpoint provides analagous functionality), two kinds of animation were used:
1) Basic slide-to-slide transition effects: I was constantly fighting the urge to use more effects. Every time I experimented with some, the video looked increasingly like a slide show with lots of random effects. You know, the kind that says: “I don’t know how to design a slide show, so I’m going to use all the effects to look sophisticated.” You know you’ve seen them.
2) Basic inter-slide animations via the Build Order window: particularly through creating a bunch of “events” (like something fading in or out, or moving its position in the slide, etc). This technique drives all the animations, such as the snake bite.
3) Videos of handwriting: One additional animation appears twice: hand-drawn images or text. For this I used Doceri on an iPad, which allows you to record drawings (like you writing on the tablet screen) and export them as movies, which you can simply insert into a Keynote slide at play it back as part of the slide presentation. You need to practice writing super duper fast!! Actually, no. Before embedding these into the slideshow, I imported these into Camtasia, increased their playback speed (it’s depressingly easy), and re-saved it as a compressed video file (to drop into a slide for playback during the slidshow). At first I really did try writing really fast to skip the additional step of speeding them up, but it didn’t take long to realize there was no possible way to write as quickly as I needed to (these are short videos; 15 seconds to watch handwriting is a bit ridiculous) and still write clearly (as I can barely write legibly at a slow speed). The speed adjustments possible through the video editing software make it easy to fit the video into the timing of the voice narration as well (more on that later).
Planning your video
You’re a good writer. You take pride in your ability to synthesize complex information and distill it into well-organized sections that make it easy to understand without being overly ruductionist. It should be easy to apply these skills to a voice-over, right? Not really, and that discrepancy makes the whole exercise worthwhile.
I had a time budget: 3 videos; 3 minutes each. Though I never would have voluntarily followed such a scheme, this was a great limitation. Ultimately, whenever I was frustrated that I couldn’t say more, it helped to remind myself the whole goal was to say less, and use the advantages of multi-media to communicate more efficiently.
My first approach was to write out a list of slides I wanted to create and what I wanted to say about them. Before actually making any slides or recording anything, I wanted to get realistic (versus wishful thinking) estimates of how long it would take to say what I thought I could say and illustrate in the alotted time.
My initial run-thrus of slides and narration that I thought would be 3 minutes were closer to 6. It was humbling, then frustrating, then infuriating, to see how many of the words on a page needed to come off the page in order to fit in the video. Everything is so important. The requiste compression helped me get very clear about what was most significant and interesting about my research in a way that even writing many versions of abstracts and summaries had not. But in this case, because both visual and audio components work simultaneously, the video is far more informative that the typical abstract or elevator speech.
Yes, the short video elides some nuance and sophistication that historians adore in writing. Please remember that people who want to bathe in your painstaking research and effortless prose will buy your book. The other 99.99999914% of the world (assuming 1000 unique readers, twice the standard academic book print run) would rather watch the 3 minute video and get the non-nuanced and simplified version; some of those will in turn want the longer, printed version. This is not about sacrificing complexity, it’s about choosing accessibility. They are not mutually exclusive.
Creating the initial voiceover
With a list of slides and a heavily-revised script, and some hope that I could stick to the time budget, I made the slides I needed, which was procedurally no different from creating any other slide presentation.
To achieve the ultimate goal of a video of the slides with my own narration, my initial plan was to play the Keynote presentation back on the screen and video capture it (via Camtasia) while talking into the microphone. I would control the slides and animations with the keyboard as one normally would during a presentation. Keynote makes it easy to record presentations—-you simply play the slideshow with (wait for it…) “Record Slideshow.”
However, this technique proved less than ideal for me. Playing or recording the slideshow with the animation effects (so, not in rehearsal mode) means Keynote necessarily and unalterably goes into full-screen mode, and there apparently no way to change that. This means, unfortunately, not having my presenter notes attached to each slide visible on the screen while the slides are playing. Why not just use printed notes? Good luck. I initially printed out my notes to avoid needing the screen, but because I ended up revising the script constantly (seriously, you have no idea), having to print out new versions each time I wanted to record was really annoying. And I started to imagine a small section of forest somewhere being bulldozed specifically for my script revisions.
The solution that worked best for me was to export the presentation from Keynote as a movie, play it back in a smaller window via my computer’s default movie player (this shows you all the animations and effects that you can’t see otherwise if you’re in rehearsal mode), and screen-capture that window as it played the video. This is super easy with Camtasia, as that’s what it was designed for. This technique is especially good if working on a large screen; at least in my case, the video (particularly animations and other transition effects) screen-captured from full screen mode was quite choppy, whereas capturing the smaller window provided much smoother playback.
It did not work out, as I initially tried, that I could simply play the slide show as it exported as a movie (in which slides automatically advance after a certain period of time), narrate over it, and keep in the narration in sync with the slides.
This was not a real problem, though, as I could simply start and stop the video to keep (or try to keep) the narration and animation in sync. Starting and stopping the movie as a way of maintaining syncronicity seems oddly lo-tech, but it’s remarkably effective.
During the recording of the narration, if you say something wrong, you can simply pause the playback, repeat what you just flubbed, then continue on with the script. It will be easy to snip out the mistake later and line up the audio and video in Camtasia. There might be times when you are trying to say something cleverly in sync with a particular animation or transition. If you mess up, you simply have to replay that part of the slide show and re-speak the lines. You’ll be screen capturing you replaying the video (all the cursor movements, etc), but it’s still easy later to snip those bits out of the screencast and put them in your blooper reel.
Using a built-in microphone is fine for rough cuts, but you’ll notice that your sound quality is infinitely higher with a some kind of external microphone to do the recording. The cheapest headset with a mic works fine; it still does a great job of minizing ambient noise, especially the sounds producted by anything connected to your computer or desk, like using the keyboard, which you’ll have to do to start and stop the video of your slides.
If you derive any joy from making things, it’s pretty exhilarating to see your “movie” start to take shape even as a super rough draft. But it’s also obvious that your initial version over won’t exactly resemble those polished videos that you might have kept in mind as a model while you worked on your own.
You can substantially improve your first rough draft really quickly. To begin, import the screencast of your slideshow playing and you narrating into Camtasia by dragging and dropping. A few minutes of experimental clicking (or Googling) will reveal how to split the video and audio tracks so that you can edit them indepedently. Any introductory tutorial on Camtasia will show you in just a few minutes how to remove sections of your video or audio you don’t want with just a few clicks and drags.
If you’ve never done any digital editing before, video editing software gives you immediately a new sense of power. It’s not always obvious, however, what the most efficient technique is. But that’s the joy of learning new things. My approach was to chomp off the obvious audio or video detritus, where I had simply repeated a passage for a cleaner articulation, or had a long pause that I needed to eliminate, or when gesticulated into my microphone, or swore at the people making noise outside of my office. Low hanging fruit, as it were.
Creating a good narration is a balance of doing it well in large chunks to maintain tone and volume continuity, and editing small bits that turn out sub-optimally. Too much editing, and the soundtrack becomes distractingly unnatural, as the volume and tone from one take won’t entirely match the next, unless of course you’ve perfected your on-demand syrupy radio voice. You can massage the volume and speed of the audio within Camtasia to some extent, but it’s sometimes easier to simply re-record a larger section so that you have less futzing and stitching to do. However much you want to continue to refine your video is only a function of your patience and time you’re willing to spend on it. Technology is hardly the limiter here.
Like any academic project, available (and even borrowed) time always seems to run our before perfection take hold, even as its always on the horizon. It was hard to be “done” with the project, despite the fact I was getting further behind in my usual work. The more you know, the more you can do, and the more you can dream up. Repeat. A little imagination and an open mind goes a long way, and reaches far more people.