So I recently got to interview a very inspirational guy, Dai Manuel, for my podcast and YouTube channel. It was a great experience with some really awesome discussion – no surprise given my subject. In addition to all the great content, I learned a ton since this was my first web-based interview, so I thought I’d share what I’ve learned.
I used Skype to do the interview, and a piece of software called “Call Recorder for Skype” from ecamm Networks that lets me record the video chat and split the two halves into their own movie files. It’s a very cool feature. This allowed me to let each side of the conversation be focused on, rather than having the screen split. It also allowed me to be off camera while Dai was talking so I didn’t have to worry about how I looked. I don’t mean whether I looked good or not, but if I had to move out of the shot for some logistical reason or something like that. It just allows the viewer to focus on the speaker more than a distracting side image.
I used a 13″ Retina MacBook Pro with a Core i7 processor, 16GB of RAM and 1TB SSD, and for audio, I used the built in mic and speakers. I was connected over WiFi, but would have preferred a hard wire. I had also run an extension chord and power strip to be sure I wasn’t on battery power. Stupidly, I turned the power strip off about 10 minutes before the call without thinking as I’m big on saving electricity. That strip also powers my treadmill, but was off, so I figured, “Hey, let’s save the planet.” It’s a habit. See how it may have mattered in the quality section.
Dai was on his iPhone, and had the Apple headset plugged in for his audio.
I also had my Canon Vixia camera running on the side, focused on the screen as a backup just in case.
Having shut off the power to my laptop, the processing power was throttled back. This ended up mattering in how well the recording was keeping up with the video chat (mainly the frame rate). If you watch the video, you’ll see some periods where there’s jagged movement. The call also dropped twice, which could have been my WiFi, the load on my computer from doing an HD recording.
Audio was interesting. Dai’s voice was very loud because he was using a headset, while mine was pretty soft. I had to adjust the audio tracks for both of us to compensate for this. Also, because I was using my built in speakers and mic, my audio track included a low-fidelity capture of Dai’s voice (what my mic picked up through my speakers). As much as I don’t want to use a headset, I will think about it strongly next time to avoid this. It wasn’t a huge problem overall, but there are moments where we’re talking simultaneously, or where the way I’ve edited things makes it awkward (he says something, then I pull my video/audio in to ask the next question, and you hear the end of something you just heard him saying). An alternative would be to use a dedicated mic, and shield it a bit so it picks up less audio from my computer.
I do think the audio from the headset – while certainly loud enough – was not as clear as I’d like. It could just be the Skype connection, but I’m guessing it was also the headset itself. It’s an OK mic, but not amazing.
running the show
I didn’t establish rules for the interview. That is, I didn’t act as a director. Dai had a lot of great things to share, and I didn’t want to cut him off. I had sent him the questions just before the interview, so he was naturally covering the material I was hoping to get into. I didn’t realize it at the time, but this lead to really long cuts of Dai talking, and less of an interview. I’m overall fine with this as I think the material is great, but it taught me a lesson for next time in case the material isn’t flowing as well or the interview is going way off topic – I should have established a gesture or something to insert myself so I could keep the interview to my intended topics and questions.
I lucked out here, and think the others I’ve reached out to for interviews will also be fine, but as I say many times, luck is not a strategy.
The split video files were really cool, and I definitely liked this approach better for such a long interview (31 minutes) than having both of us sharing the screen. With the audio issue I mentioned above, I don’t think it would have worked to have it all simultaneous since everything Dai said would come with the version my mic picked up layered over it. While you don’t have to split the video (you get a combined file, and then have drag-and-drop filters you plop the file on to create the split versions), you have to say you want this upfront as the output file is different (it’s essentially a 32:9 file made of two 16:9 images side by side versus creating a single 16:9 image that’s split between an 8:9 of me and 8:9 of Dai). That means I couldn’t really opt to go back to a side by side. So I was committed to it.
Luckily, ecamm built in a droplet to split out audio sides separately from video sides (you can get both together, or separately). There are times when I wanted to clean up the audio tracks a bit, and needed them separated from the video. These filters alone made Call Recorder worth the price (it was $29.99).
The downside to this approach is that I ended up with lots of back and forth between us. I could just cut back and forth, but it felt too choppy, so I used transitions. Every time I use a transition, I’m reminded of a Simpsons episode where Homer makes a dating video of Ned Flanders using a kid’s video editing board Lisa has. He uses a star-wipe effect between each cut scene, and Lisa warns him that it’s too much. He pushes back, insisting you can never have too much star wiping. Lisa was right, but the video is hilarious. Anyway, that always makes me concerned I’ve overdone it with the jazzy transitions. I scaled back to really simple fades most of the time as the novelty wore off and I ended up almost feeling nauseous from all the cubed-wiping and card flipping going on as Dai and I went back and forth.
editing takes time…lots of time
I spent two days editing the 31 minute final product. Part of that was time I lost because of a computer problem, but a lot of it was nitpicking trimming clips, aligning audio, and trying to get nearly 50 minutes of content down without losing the point. I honestly wouldn’t have cut anything, but it was too long for most viewers. So if I was going to cut, it had to be well done and the essence of everything Dai was saying had to be intact.
in the end…
I think the content is what made the interview solid. Dai’s an amazing guy, and really gave of himself in a very real way in the interview, and you can’t help but be inspired coming out of it. I was inspired myself by his words, and also by what I’d do differently next time. In summary:
- make sure my laptop is plugged in, and all other programs are quit
- establish rules of engagement so I can be the director and get the interview I’m looking for
- find a better way to capture audio on both ends (in case the next person uses a setup like I had), or ensure we don’t talk over each other
- make sure I have solid editing time blocked out
- make sure your interviewee is willing to help promote the video and earn the right to check in to make sure they do