What is Your Podcasting Set-up?
I answered this question a while back elsewhere in the forums, but thought it would be better placed in the FAQ section ... so here it is, with a few updates.
We have two set-ups really; one when Mark and I do a double-ender (most of the time), and one when Mark is visiting and we're in the same room.
For the double-ender ... We have our "conversation" on the phone. Mark is recording his end of the show using a Perception 100 microphone with a Marantz PMD-670 portable digital recorder. Given Mark's hectic travel schedule, this is a *great* set-up. Not only can we record during his travels, but we've used it effectively with clients as well.
I record my end using a MXL 990 microphone into a Alesis Multimix FX (non-USB) mixer into my PC sound card. I record (and edit) using Adobe Audition. On my side, I record MY VOICE on the left channel, and the phone audio (Mark) into the right channel. More on why (i.e., what's a double-ender) in a second ...
When we're done recording, Mark uploads his end (WAV file) onto the web server, and I subsequently download to my computer.
Then I import both tracks into Audacity and align the two tracks. This, by the way, can be a real pain. I couldn't figure out why it was so difficult until I had coffee with Doug Kaye of IT Conversations. I didn't know that, because of different clock speeds in each computer, even if you synch up the beginning of the two files, over time they'll drift apart. Over a 40 minute show, it can be as much as 5-10 seconds apart! So ... one must "synch-up" the different tracks *several* times over the course of the show.
To solve this problem, I pull both my recording and his recording into Audition. I then align MY right channel (the low-quality sound of Mark as it came through the phone) with his high-quality recording from the PMD670. Then I delete my right channel (low-quality Mark ... errrr, I mean low-quality Mark *audio*). I'm then left with two high-quality recordings of both of us.
Once I've done this, I do some RMS normalization on the files, adjust volumes so they are generally equal and clean it up sufficiently so that it's easier to edit.
Then, I do all the editing ... take out, as much as possible, kids running around upstairs, creaky chair, heating kicking in, etc. (Hint: turn off air conditioning or heating before starting to record. Second Hint: Remember to turn it back *on* when you finish late at night in the middle of winter ... improves marriage considerably ... I learned the hard way).
Then I record intros, outtros, etc. and run through same RMS leveling process. Then put all the tracks together in Audition, add music, align everything ... and finally export as MP3.
One additional note: I've recently found that the quality of the MP3 is *much* better if I use the command-line version of LAME vs. using Audition to convert to MP3. I get more control of the process and there are a couple of things one can do to improve the quality (using a lowpass filter, for example). So after exporting out of Audobe as a WAV file, the LAME settings I use are something like this:
lame -q 0 --lowpass 10 --resample 44 -b 64 ./manager-tools-2006-03-06-WAV.wav ./manager-tools-2006-03-06.mp3
Then I used MP3tag to add all the MP3 tags and upload to the servers. Create the show notes and link to the mp3 file and I'm done.
When we record together, same process except Mark's mic goes directly into the mixer and no synching up of our voices required.
How long does it take? Once the recording is done, the production work is *generally* 2 to 3 hours. Occasionally, there are sound quality issues (e.g., background noise) that require some cleaning up (I used SoundSoap for this) that increases the time significantly.
One additional note: if you do a lot of these, be sure to get a LARGE hard drive. And if you plan on having backups of all your hard work, plan on getting TWO large hard drives. ;-)