Subtitle: How not to do a podcast.
Well, TUAW now has two podcasts under our belt (link to one; link to two), with many more to unfurl in the months ahead, I'm sure. FYI, we're scheduling these roughly at once a month.
The first podcast featured me talking into an AKG mic that I had plugged into a Roland UA-30 USB. I used GarageBand to record my audio, setting the audio input to the Roland UA-30 in Garageband's preferences, and then creating a new track by going to Track—>New Track, then selecting Real Instrument, Vocals, and No Effects. I paused often, edited quite a bit, and shoved a few audio clips in in their own tracks by simply dragging them into GarageBand. Then I exported to iTunes, and used iTunes to convert to MP3. Then I emailed it to the Weblogs, Inc. elves who slaved in the podcast mines for three hours. Then I posted the link.
Read more after the jump...What was wrong with it? Too quiet and only me talking. Too many "ums."
For the next podcast, Scott and I started up a Skype conversation. I used a Logitech USB headset that I bought specifically for Skyping with friends worldwide. Scott used his iSight as a mic. I set my headset to be the input for Skype and the Microphone source for GarageBand. Scott set his iSight to be the mic for both Skype and for Quicktime. I began recording in GarageBand. Scott launched QuickTime Pro and chose File-->New Audio Recording and began recording. I said, "Okay, repeat after me: One, two, three." And Scott replied, "One, two, three." Then we had a very long conversation and whenever we had to pause or stop, we would start back saying 1 2 3. This becomes important in the next step.
After we were done, Scott exported his Quicktime Audio movies to WAV files and then dropped them into iTunes and converted them into MP3s. He then sent them to me via iChat. I dropped them into different tracks in Garageband and used our 1, 2, 3 to sync everything up nicely. I then went through and carefully edited bits. Whenever I cut bits, I made sure to split the audio on both my track and his track, so that I could move everything down the line together and keep things in sync. This took a good hour.
After I was done, I exported to iTunes and then converted to MP3.
What was wrong this time? Well, the volume was up too high on my headphones, so while Scott and I were Skyping, his voice was being picked up slightly on my headset's mic. Thus the faint buzzing you hear throughout the podcast. Also, this time I made everything much too loud, trying to overcompensate for the quietness of the first podcast.
I had the MP3 compression set to much too high a rate, and as a result our podcast was a meaty 20MBs in size. Since then, I've re-compressed the file and it's down to half the size.
This seemed to work very well for getting Scott and I both sounding clear. Much more effective than the record-the-iChat / Skype audio podcasts I've heard around the web.
Now to figure out how we're going to manage our next podcast with 5 TUAWers chatting...
Reader Comments (Page 1 of 1)
6-15-2005 @ 11:55AM
mike said...
Nice post, great comments.
Thanks for sharing, everyone!
-Mike
http://www.GarageSpin.com
Reply
6-16-2005 @ 4:20PM
Kristoffer said...
You could use ichat to have an audio-conference between you bloggers, and record all of it with Audio Hijack (I think that would work). Just a tip!
Reply
6-16-2005 @ 4:20PM
Twist said...
I haven't listened to any of the pods casts here but your method does sound like it would produce a high quality one. Should have a more of a sitting in a booth together feeling to the listeners than the Skype or iChat captured methods. Downside of course is all the extra work.
Reply
6-16-2005 @ 4:20PM
C.K. Sample, III said...
Hey Kristopher,
Yeah, Scott and I tried something similar with WireTap Pro, but it crapped out on us, and the over iChat / Skype quality of Scott's part of the feed sounded like crap too. We got good clean sound by grabbing it separately, and I like that. No internet-interference.
Reply
6-16-2005 @ 4:20PM
Jack said...
Regarding MP3 compression, I highly recommend circumventing iTunes native MP3 compression and using the LAME MP3 encoder. You can do smaller sized files at lower bit rates that sound amazing with it. Here are the basic steps to install LAME and using it with iTunes:
1) FIrst, download the LAME 3.96 package and install it:
http://www.thalictrum.com/index.php?pageid=2
2) Next, download 'iTunes-LAME':
http://blacktree.com/apps/index.php
This will bridge the CLI interface of Lame with iTunes.
3) Once installed, just launch iTunes go to the script icon menu and choose "Import with LAME..."
A floating window will open up that will allow you to set LAME settings and run it simply in a way that will import the MP3s into iTunes seamlessly.
The process is easy:
1) Create a playlist for the MP3s you want to convert.
2) Drop the files you want to convert into the playlist.
3) With the playlist selected launch 'iTunes-Lame'.
4) Choose the preset settings or enter your own.
5) Hit the 'Import' button and wait.
Depending on CPU speed, the process can be longer than standard iTunes imports. But once you see the smaller size of your files and listen to the quality of the MP3, you will never turn back.
And here are the two custom LAME settings I use regularly.
Small But Great Sounding Stereo MP3s:
--lowpass 19.7 -V3 --vbr-new -q0 -b96 --scale 0.99 --athaa-sensitivity 1
I use this all the time for CD importing or MP3 conversion. The space saved is amazing and I can't tell the difference between this tweaked VBR setting and other higher bitrate VBR imports.
Small Mono Files for Radio:
--resample 8 -V3 --vbr-new -q0 -B16 --lowpass 15.4 --athaa-sensitivity 1
I have a nice sized collection of old-time radio shows I have encoded and reencoded using this setting. Perfect for old AM radio shows. Or a smaller sized version of a radio show you might want to post for people who are not particular about having it in stereo. There is defintely a loss of high-end with this setting, but there is little to no distortion and unless you have heard the original, you will barely know the difference with this setting in many cases. Great for mono radio shows or old-time shows.
Reply
6-16-2005 @ 4:20PM
Jer said...
Thanks for the nice write-up.
Back in the days before timecode the clapboards you see in movie making (ie. scene 1, take 12) would be forcibly snapped shut. The clap makes a very sharp spike in the waveform which is easier to line up with that point in the film where the clapboard was closed.
Reply
6-19-2005 @ 3:58AM
robert Walch said...
My podcast is a podcast about interviewing other podcasts. I do all the interviews with skype. I found the best way to record was to take the audio out from Skype into a Mixer with the audio from the Mic being split from an M-audio Mobile-pre USB to the Mixer via XLR and to the Computer via USB. Then take the mixed signal and record to an iRiver at the 44 Khz Sample rate and 160 kBps Mono bit rate.
This method can be used to record 2, 3, 4 or 5 people on skpe. Since it is all recorded on your end. Also gets rid of the for the double ender.
Here is a link to my interview with Adam Curry.
http://www.podcast411.com/podcasts/411_050601.mp3
If you want the block diagram for this set-up send me an email to
rob@podcast411.com
Rob W
podCast411
www.podcast411.com
Reply
6-19-2005 @ 3:59AM
robert Walch said...
My podcast is a podcast about interviewing other podcasts. I do all the interviews with skype. I found the best way to record was to take the audio out from Skype into a Mixer with the audio from the Mic being split from an M-audio Mobile-pre USB to the Mixer via XLR and to the Computer via USB. Then take the mixed signal and record to an iRiver at the 44 Khz Sample rate and 160 kBps Mono bit rate.
This method can be used to record 2, 3, 4 or 5 people on skpe. Since it is all recorded on your end. Also gets rid of the for the double ender.
Here is a link to my interview with Adam Curry.
http://www.podcast411.com/podcasts/411_050601.mp3
If you want the block diagram for this set-up send me an email to
rob@podcast411.com
Rob W
podCast411
www.podcast411.com
Reply