A few months ago, I started recording a podcast with some friends. At that time, our process was to gather everyone into one Skype call and record the session with WireTap Studio. That worked reasonably well, but there were a few problems.
The quality of the end result was dependent on everyone's Skype connection. Throw in a bad rainstorm, a large download or an iffy wi-fi connection and quality suffered. The other issue was people talking over each other. Even the most disciplined podcasters do it occasionally. Since the raw audio was a single track, editing the overtalking out of the final show was difficult, if not impossible.
A few episodes into it, we read Dan Benjamin's tips on recording a podcast with participants in multiple locations. He and John Gruber use a time-tested method called a "double-ender" when recording The Talk Show. As they talk via Skype, John and Dan record their own audio locally, and those two files are edited together as separate tracks. We've adopted a similar method using GarageBand, and it's been working wonderfully. Read on for details on our setup.
First, launch GarageBand and connect your mic. I use a Logitech ClearChat Pro USB Headset mic that cost me about $35US. Hardly a pro-level solution, but it works well (plus, I'm hoping to find something a little better under the Christmas tree in a couple of weeks).
Next, ensure that GarageBand is set to record from your mic. I've found that I get the best results by recording from my mic directly into GarageBand. From the preferences pane, click Audio/MIDI and select your mic as both audio input and output if it has built-in headphones. Avoid using external speakers for audio out, or the recording will seem to have an echo as it records your voice once from your mouth and again from the speakers.
Finally, click the track you're going to use and ensure that the button below that track's title is red, meaning recording is enabled. Once that's set up, I call the others on Skype, and they prep their copies of GarageBand in the same way. From there, we all begin recording and say, "One, two, three, go" together. Believe me, that simple step makes editing a heck of a lot easier.
That's really it. We each record our own audio while Skype simply allows us to talk together. When we're done, each participant exports his or her own track and sends it to whoever has editing duty. When that's me, I make one track for each person, line them up according to the countdown and begin editing. Finally, I add a jingles track for the intro music and any additional sound effects.
This way, the quality of the Skype call doesn't affect the final product, and instances of people talking over each other are easily corrected, as each voice is its own track.