I've been recording my presentations for about a year and a half now. There are a couple of reasons why I've wanted to record myself speaking:
- By watching myself, I can learn my quirks that I don't notice. This helps me improve my speaking technique.
- I can put together clips of me speaking to create a promo video to help get me more speaking opportunities. (I never did this because I had some great professionally recorded videos to showcase before I got around to completing it.)
- I can post recordings that come out well to my YouTube channel.
And this is from March 2015 (Nebraska.Code(), Lincoln NE):
So here's what I've done. It's not professional quality, but it gets the job done for what I was looking for.
Sound Quality is King
One of the most important things about recording live presentations is making sure that you have good audio. If there is echo-y sound that's hard to understand, no one will watch the video. I've passed up watching videos of people I really wanted to hear just because the audio was so distracting. This problem actually kept me from recording my presentations for quite a while.
For a previous discussion on the importance of audio quality, check out my article on screencasting: Jeremy's Screencast Production.
Record Audio Separately
Don't rely on the camera or your computer to pick up the audio. Record the audio separately. I use a Zoom H1 digital recorder (Amazon link) that I'm very happy with:
This has a built-in microphone array which would be good for recording interviews or something where you don't move around a lot. For my presentations, I pair this with a Sony ECMSC3 lapel microphone (Amazon link):
This is an inexpensive microphone, but the quality is surprisingly good. I can clip the microphone to my collar and stick the Zoom recorder in my pocket.
Note: If you take this approach, make sure the microphone is plugged in all the way to the recorder. I have a couple presentations that are nothing but a recording of the inside of my pocket.
What I Like
I am happy with the results of the Zoom recorder. It runs on AA batteries and records to a micro-SD card. Battery life is very good, and 2-1/2 hours of audio takes about 1-1/2 gigs of space (which isn't much when you can buy 64 gig cards really cheaply).
I use the "Auto Level" feature of the Zoom recorder (you can see the switches on the back). This has a few quirks, but it's easy to fix in editing (more on that below).
You'll also notice that there is a "Hold" switch. This will save you from accidental button presses, particularly if you put the recorder in your pocket.
Repeat the Question
A general piece of speaker advice is to always repeat the question so that other members of the audience can hear it. This is especially important when recording because you will not pick up the audience at all on the lapel microphone. So you want to repeat the question for the recording as well.
I use Camtasia Recorder (from TechSmith) for screen capture. I've spent a bit of time talking about Camtasia in my screencasting article, so I won't repeat it here. I'll have a bit more to say about it when we get to editing.
One thing to be aware of is that the Camtasia Recorder will capture some keystrokes for its own use. So be prepared to re-map some keys in Camtasia or be ready to use alternate methods in your demo. The most prominent one that I run into is that F10 is "pause" for Camtasia, so in Visual Studio, I'll use the toolbar buttons for debugging instead of F10 to step through code.
High-quality video was not really important to me. I primarily wanted to be able to see myself in action (and again, potentially put together a promo video). Rather than spend a lot of money on a video camera, I spent a couple hundred dollars on a little Sony HDR-AS100V (Amazon link):
This is the Sony version of a GoPro. It's really designed as an action camera, so the lens is a bit fish-eyed. I got this a while back, so I'm sure there's an improved version of this now.
The recording quality is good: 1080p. And if you crop the video (like is shown in the first video in this article), then you don't see the fish-eye. Note: the second video in the article was actually recorded first, and I learned some things about editing that I was able to use in the other video. But more on editing later.
I have the camera mounted on a small tripod (this was included with the Zoom kit that is linked above):
It's nice and small to stick in a bag with the other gear. But the legs expand to give it a decent height when placed on a table. Here's the rig with the camera attached:
I usually put the camera on a table in the front row (which gives a bit of a side-angle). I've put it in the back of the room a few times, and the image was a bit small. When it's closer, it's easier to see my movement and mannerisms.
What to Watch Out For
There are a couple things about this camera to watch out for. First, it can only be charged by the USB port (there's a removable battery, but you would have to buy a separate charger for that). It cannot be plugged in and record at the same time, so you're limited by battery life.
In my experience, battery will go about 1 hour 45 minutes. This can usually accommodate one of my presentations at a user group, but I make sure to start it just before I start, and I'll even pause it at breaks to save a bit of battery.
This also uses micro-SD for storage. And it runs about 4 gigs for 35 minutes of video. I use those weird increments because it actually splits up video files into 35 minute chunks. Again, with a 64 gig card, I've never had any problems with space. The limiting factor has always been battery life.
Record audio in all three places: digital recorder, screen capture, and video camera. Even though the final audio will be from the digital recorder, it's easy to use the sound waves to line up the various tracks when editing the video.
And here's a tip I picked up from my friend and fellow speaker Justin James (@digitaldrummerj):
After starting all 3 recorders, clap your hands.This will create a very visible audio spike that you can use to line up the various tracks. (Just like movies use the clappers to start a take.)
Lots of Files
The result of the recordings are lots of files:
The ".mp4" files come from the video camera. As mentioned, this records in 35 minute chunks, so these need to be put together in editing. The ".trec" file is the screen capture from Camtasia. And the ".wav" file comes from the Zoom audio recorder.
The ".camproj" file is the Camtasia project. We're headed there next.
Putting Everything Together
Now that we've got all of the sources, we need to put everything together in Camtasia Studio. First job is laying down the tracks and lining them up. This is where the audio helps A LOT.
First, here's a look at the Camtasia interface:
The order of the tracks are important, the track that is "on top" (in this case, Track 4) appears over any other tracks.
Track 1: Audio track (not visible). This comes from the Zoom digital recorder and it will be the audio that we ultimately use.
Track 2: Screen capture (left side of screen). This comes from the Camtasia Recorder.
Track 3: Video capture (right side of screen). This is from the Sony video camera.
Track 4: Overlays (text at top of screen). We'll take a closer look at this a bit later.
Lining Things Up
After getting the media in the right slots, we have to line everything up. For this, we can look at the audio for each track:
We can use the audio tracks to line everything up. Notice that the peaks line up across the media. And as noted earlier, a "clap" at the start can create a spike to make it easy to see where the start point is.
In this case, I've already trimmed down the video and audio, so we can't see that part. Once everything is lined up, you can trim the excess at the beginning and end of the tracks.
You'll see different intensities when looking at the audio, this is because the computer is one side of the room, the camera is on the opposite side, and I'm walking around with the voice recorder in the middle.
After getting all of the tracks lined up, we can use the Audio tab in Camtasia to simply mute the 2 video tracks. This will leave the audio just from the Zoom recorder.
When I'm doing screencasting, I don't normally use the audio leveling tool (as mentioned previously). This is because screencast recording is in a much more controlled environment where I'm sitting still. Things are a bit different when I'm doing things live and don't get a "second take".
So I use the volume leveling tool in Camtasia:
Just select the audio track and click "Enable volume leveling". This smooths out the peaks, and the audio track now looks more like this:
This is a bit different from the audio track we saw above.
I do this to smooth out some of the "Auto Level" from the Zoom recorder (remember I mentioned that above?). The auto-leveling of the Zoom recorder generally works well, but if you cough, tap the microphone, or cause some other "spike", the recorder will lower the sound levels, and it takes a few seconds to re-adjust.
The "Auto Level" in Camtasia will correct this. It's not ideal. But I also don't have a professional audio technician doing these recordings for me. The audio still comes out a lot better than many of the "live recordings" that I come across.
There are a couple different approaches to take when trying to show both the screen and the person at the same time. For the April 2016 video, things were pretty easy:
This projector was had a 4x3 aspect ratio, so my screen capture had that same ratio. This made it really easy to put the screen and video side-by-side (although it does look a little strange in the beginning when I'm showing my wide-screen slides).
In contrast, in the March 2015 video, I put in a lot of edits to show the live video when that was important and to show the screen when that was important. This was *a lot* of work. It is much easier to do side-by-side or an overlay if you can get away with it.
Since the videos were side-by-side, the order of the tracks wasn't as important (since they don't overlap). But if you're overlaying or doing other interesting things, you want to make sure that the right video is "on top".
As mentioned above, the Sony video camera records in 35 minute chunks, and these end up as separate files. The good news is that if you put the two media files right next to each other in Camtasia (on the same track and touching each other), there is no gap or stutter at all.
Cropping and Panning
It's possible to crop and pan inside Camtasia to show just parts of the video. Some of these features are really hard to find. For example, the "crop" button is in the top right corner of the preview window:
This is entirely non-obvious, and I haven't found the "crop" feature on another screen or menu (maybe I'm missing it).
Here's what the video looks like un-cropped:
So you can see that I cropped out quite a bit of the frame. Cropping also can make the "fish-eye" of the camera a bit less obvious. This is why having the 1080p recording is nice. Even if you don't show the video full screen, you can crop and zoom into different areas and still have decent quality output.
This video was shot at the Central California .NET User Group in Fresno, CA. It's a small group; there were 10 people there that night. But it's not always the size of the room that makes a good presentation. Sometimes the smaller groups are better because you get more questions and interaction. (And remember to "repeat the question" even in a small room. Otherwise, you'll have an audio gap in your recording.)
The last step in the editing process is to add callouts to the video. I've talking about adding callouts in screencasts, but it's also important when showing live presentations.
Here's a good example of why:
You'll notice in the video that I'm pointing at the screen. This is pretty common for me (especially when I'm talking in a training room). But it's impossible to tell what I'm pointing at.
This is where the "Callouts" feature in Camtasia comes in handy. I can highlight the area on the screen that I'm talking about. In this case, I'm pointing to a method signature and using a red rectangle to highlight it. This is on the "top track" (Track 4 here) so that it shows up over everything else. I've added similar callouts throughout the presentation.
After getting all the bits together, it's time to render and upload. I try to render at the highest resolution I can (which is generally 1080p). Then when I upload to YouTube, they'll process it down to lower-quality versions.
Rendering projects with live video take significantly longer than rendering screencasts. So be prepared for that. This is compounded because my screencasts are generally in 20 minute chunks and will render in about 20 minutes. But a live presentation is over an hour long, so it's both more material and also a slower render.
Make sure that your computer is well ventilated during this process.
This is a lot of work. But it's not that hard to do. The equipment cost me around $500 (which isn't cheap), but I've gotten quite a bit of use out of it. The Zoom recorder in particular has been a great little tool. I'm surprised at the quality that I can get from this little plastic device. This is the recorder that I also used to record bits of my banjo playing.
I have recorded about 20 of my presentations. And I've gone through the process of laying down the tracks, lining up the audio, and rendering each of them. Most of them have just been for my personal use (so you don't see all of the work behind those). But a couple have made it out into the wild, and when I come across a good presentation, I'll be sure to post it.
If you decide to start recording and producing your own presentations, I hope that my experience will make the path a bit easier for you.