By Jason Snell
January 14, 2015 1:16 PM PT
How I podcast: Recording
I do a lot of podcasting. And I am often asked about what tools I use and how I produce my podcasts. So in a series of articles on this site, I hope to detail my approach to making podcasts. What I don’t intend is to suggest that this is the only way to make podcasts—it’s just the way that I make them. If I can provide some sort of inspiration—or even a cautionary example of what not to do—I’m glad to do so.
While I think it’s true that many people underestimate how much work goes into making a podcast, I also get the sense that other people overestimate the time I spend. And depending on what kind of a podcast you’re creating, the amount of time required to put it together can vary widely. The average episode of The Incomparable probably takes three or four hours to edit; the average TV Talk Machine I can turn around in 10 minutes.
Equipment and Recording
To get started, I’ll refer you to this post, in which I suggest that you can get started with podcasting with minimal equipment—what you really need is a passion for what you’re going to be talking about on your podcast.
I’m a big fan of the Blue Yeti microphone, which is often available on sale for $75 or so, though as I write this it’s going for $100 on Amazon. There are lots of other options. If you’re going to be podcasting using Skype, you should also use headphones—and ideally they’d either be in-ear headphones or closed, over-the-ear headphones. Earbuds are less good because they leak noise—but again, if earbuds are all you have, so be it.
Over time I accumulated a tripod, a shock mount, a pop filter, and lots of other gear, but I did probably 100 episodes of The Incomparable with the Yeti sitting on my stomach. Low tech, but it worked.
I have recorded podcasts in a wide variety of environments, but the most common one is a quiet room, with the door closed. The less echoey your room, the better. Bookshelves, curtains, carpet, and similar stuff on the walls and floor are helpful in reducing echoes. I wish I had more tips about reducing echo, but I’ve been fortunate to mostly record in rooms that weren’t echoey to begin with.
Most of my podcasts are recorded via Skype, so I’m the only person in the room. If you’re recording stuff with multiple people in the same room, that’s got a very different set of audio requirements that I’m not going to cover—at least not right now.
Everyone hates Skype
Most podcasters I know use Skype to record their sessions. Everyone hates Skype, because it can sometimes (occasionally? frequently?) behave strangely and make it difficult to hear what other people are saying. But everyone still uses it, because it’s the best option out there, flawed as it is. Skype lets you have audio conversations with quality beyond what you’ll find on a regular phone call, and you can have calls with two, three, five, even ten people on it all at once. (Things get insane with that many people—I don’t recommend it.)
There are two basic approaches to recording with Skype: You can record the conversation as you receive it over the Internet from Skype, or everyone on the podcast can record their own end of the conversation. Believe it or not, I’ve seen podcasters hotly debate these two approaches. But podcasters do like to talk, so is that any surprise?
When Skype is working well, it almost sounds like the person is in the room with you. Skype’s audio algorithm squelches background noise and levels out the volume, too. If your connection is solid and you’re recording a two-person conversation, just recording the audio from Skype and using that as the basis for your podcast is completely viable. I’ve been doing this for TV Talk Machine, and it’s worked well.
But if your Skype connection should falter—and Skype connections, they do falter—you can end up with something that’s a mess of artifacts, silent patches, and other bad sound. If you can’t get a clear, reliable Skype connection, I’d advise against using this method. And if you’re having a conversation with more than two people, the Skype thing gets ugly in a hurry.
Recording a multi-ender
The method I use for every other podcast I do is this: Everyone records their own microphone and sends me the files when we’re done. While it definitely ups the difficulty level a bit, the results are usually worth it. In the audio game this is called a “multi-ender,” and it requires some work from the people on the other end of the line.
Fortunately, if they’re Mac users, it’s pretty easy for them to record their end of the conversation. The QuickTime Player app, included with every Mac, lets you record your own audio. After launching QuickTime Player, choose New Audio Recording from the File menu. In the resulting window, click on the drop-down triangle next to the big red button and select your USB microphone of choice, while ensuring that quality is set to High. Then you press the big red button to start recording. When done, press the button again and save the file as a QuickTime movie and send the file to whomever is editing the podcast using Dropbox or some other file-transfer service 1. (For Windows users the workflow is slightly more complicated, but they can download the free Audacity and record using that app 2.)
Another approach—and the one that I use—is to record conversations via the $30 Call Recorder for Skype by Ecamm Software. This plug-in for Skype records your own microphone input and the output of your Skype call as separate tracks in a single QuickTime movie. (An included utility app allows me to split that file into separate files for each track and convert them to various other formats.)
If the other people on your Skype call are using Call Recorder, it also eliminates a point of failure: Skype conveniently allows you to set separate audio input and outputs from the defaults used by your Mac. This can be awfully convenient, since you can set Skype to always use your microphone and headphones, no matter what the input and output are set to in the Sound preference pane. On more than one occasion I’ve had a podcaster who sounded fantastic on Skype send me a file that was accidentally recorded from their laptop microphone or EarPods or some other source that wasn’t their good microphone 3. If a podcaster uses Call Recorder, I have 100 percent confidence that the microphone I’m hearing on the call is the microphone that’s being recorded, because Call Recorder records whatever Skype sends.
I’ve had people use other utilities, such as Piezo and Audio Hijack Pro, to record their conversations. These can be great tools, but their added complexity also increases the chance of user error. Recordings always get made, but often the user has misconfigured the software so it’s recording the wrong thing. Sometimes it’s a recording of Skype, or of Skype with the user’s own voice superimposed, or it’s the wrong microphone. That’s why I generally recommend that new users stick with the free QuickTime, and more committed podcaster use Call Recorder or (in the case of someone using Windows), Audacity.
The best thing about the multi-ended recording session is that every participant has a local recording, so it’s of a higher quality than anything that could be sent over Skype. The second best thing is that every participant in a multi-ender has their own, isolated audio track. That gives you a lot of power when it comes time to edit the podcast.
When everyone records their own end, there’s some extra work on the editing side. First, you need to use a multi-track audio program, though GarageBand comes free with every Mac and it’ll do the job. Second, you need to line up all the files. That can take some time. Many podcasts I’m on ask people to start recording and then count down from three and clap in unison, which gives an editor a visual sign to use to synchronize the tracks. After more than a year of editing The Incomparable using a similar approach, I realized that while everyone pressed Record at different times, everyone stopped recording pretty much at the same time — at the very end of the call. Once I started lining up recordings so that they matched at the end, rather than the start, my job got a lot easier 4.
Having a recording of your Skype session is incredibly valuable when it comes to syncing up multiple audio tracks. Different computers can record audio at slightly different rates, so on long recordings you may need to split an audio clip into smaller bits and slide it around a few times in order to get it to line up from beginning to end, always using the Skype recording for reference.
Once you’ve got the tracks recorded, it’s time to edit! Editing will be the topic of my next post on this subject, so stay tuned for that. Any questions? Ask @jsnell on Twitter or email me at jsnell at sixcolors.com.
Anze Tomic adds, “When it’s time for the guest to send me the file, if they are not a Dropbox user, by far the easiest thing is to guide them through how to send the file via wetransfer.com. It’s almost like sending an email, so most people can grasp it right away.” ↩
Anze Tomic writes, “When we finish I ask the guests to export the audio form audacity in the ogg vorbis format since I know I’ll be able to open that right away on my end. (Audacity can do that out of the box, so there is no need to install an mp3 codec and it saves on space since an hour long .wav file is huge.)” ↩
In those cases, I generally fall back to my recording of the Skype conversation, which offers their best quality audio despite it having come via Skype, but it’s a lot of extra work. ↩
For the last year or so I’ve been using a command-line app to synchronize my tracks, and it’s great. Unfortunately, it’s still in private beta. When it’s finally public, I promise to shout it far and wide. ↩
[If you appreciate articles like this one, help us continue doing Six Colors (and get some fun benefits) by becoming a Six Colors subscriber.]
[Don't miss all our podcasting articles.]