What goes wrong in recording audioblogs…

Yesterday I reached a minor milestone: the tenth of my audioblog entries, and I have a strong suspicion I may have had more than ten downloads. I also chose to wrote up a brief description about how I make these entries, and I hope that some people found it useful. After another minor glitch in yesterday’s recordings, I thought it might be useful to also write up all the problems that I have. You can consider this a set of gotchas to watch out for, and also as a wishlist for creating tools that make recording these shows less error prone. Toward the end, I give some advice which I try to obey myself as I go into the future of audio weblogging.

Mixer Dialog!Sound card configuration: I mentioned this a bit yesterday in the text blog, but I reiterate with a bit more detail. On the Mac, many people use WireTap, but on my PC I have found that many soundcards have the ability to record directly from their internal mixer. If your lucky, when you open it and look under the Options > Properties > Recording dialog for devices, you get something that looks like the image on the right. You generally want to record from the Stereo Mixer, not the microphone. Then anything you play back through PC should get recorded by your recording application.

This was actually pretty straightforward to setup on my laptop, but my new Media PC has 7.1 sound, which means that instead of a single sound card, it looks to WinXP as if there are four or five separate devices, and some of these devices reconfigure themselves automatically when speakers and headphones and microphones get plugged in. This results in a larger space of possible settings to search, and some of them inexplicably don’t work. I’m still trying to figure out the rhyme and reason of the whole thing myself, so don’t be discouraged if you can’t seem to make it work right off the bat.

Levels: Of course what invariably happens is that even when you get the recording devices setup, the levels and loudness of things are completely inappropriate. The single thing you want to do is avoid clipping: I suspect that the artifacts in my closing music yesterday were due to the final clip being too hot, and clipping off. Sometimes minor clipping isn’t bad, but it can introduce both popping and when you reencode at fairly low bitrates, all sorts of bizarre artifacts. I try to monitor the audio levels using Audacities record meter, and often jigger the recording levels on the fly. It’s most convenient to do that using the mixer, which you will want to leave open anyway so that you can mute the microphone (yes, I didn’t mute the microphone during one of my openings a couple of days ago, and you can hear me thumping around). I usually set the record levels at their midpoint setting, and then manipulate the playback levels during the “performance”.

What is needed: Both of the above problems could be ameliorated by an audio recording wizard followed by a profile manager. You’d like the smarts of audio configuration to be encapsulated so that it would automatically configure the recording and playback settings and their levels, and then be able to save and restore these settings all at once. Then you wouldn’t have to go through the error prone setup each time you changed your settings for some other purpose.

Try to have a plan: I don’t script my audioblog entries (could you tell?) but I do find it useful to generate a short checklist of the major points I want to cover. When I don’t have a plan, I panic as I try to fill dead air, and my speech becomes filled with all the classic irritations.

Resist the urge to over edit: I have pulled out short problems in my recordings as a courtesy to my listeners, but you learn more by mistakes than you do by trying to produce “the perfect recording”. I find even the worst audioblogs to be better than 90% of commercial radio, if only because people are always talking about something they are passionate about. Don’t bother trying to be too professional: it is the professional media that has bored us into trying broadcasting for ourselves.

That being said, try to improve: I listen to my audioblogs probably more than anyone else, and I try to do so critically to see if there are systematic things I can do to improve the quality of both the technology and the presentation. I also listen to other audioblogs to get ideas for things that are working for others, and try to adopt them into my own presentations.

Don’t worry! The worst thing that can happen? Nobody listens to you. But nobody was listening to you before, so you haven’t lost anything. 🙂