A Unitarian Universalist Welcoming Congregation

About Our Podcast

by ucmhadmin
Wednesday December 29, 2010 at 4:39 pm

by Ben Peck


I’ve been a web developer for the past 10 years, and have also had experience with recording audio as a hobby. Bringing the two together was very natural, but after putting together this explanation, I’ve realized that this whole process is quite complicated! There are much easier ways to produce a simple podcast, but I am striving to make our PodCast (and our website) as professional as possible.


Equipment and Setup

Our church has a permanently set-up PA, mic and speakers. The PA has a single mono output. So, we purchased an iRiver T30 (at Amazon for around $130), a stereo 1/8th” to stereo rca adapter and an rca cable. We take the mono PA output and record it to one channel (left or right, doesn’t matter). On the other channel we attach an Archos Jukebox Stereo Microphone connected in mono, for just the choir. At this point it isn’t ’stereo’, but rather 2 channels. When I later import the audio into the computer I split it into 2 mono tracks.


The iRiver T30 is a 1GB mp3 player with recording. iRiver is the one brand of mp3 players that has a line-input and a built in mic, and has several options for recording. I’ve set the options in the iriver to do the highest quality recording possible (which equates to 44.1 khz @ 340 kbit for line in, 128 kbit for built in mic). Since we don’t use the mp3 player for anything else, it has plenty of memory to do several recordings before needing to be emptied.


About 5 minutes before the service, we turn on the PA and start the iRiver recording. At the same time I start a stop-watch. As the service progresses, I write down the times that things occur in my order of service – this helps with the editing process later.


When the service is over we stop the recording and turn off the PA. We enjoy our coffee and treats, then I take the iRiver home (although I’m tempted to bring my laptop in, as our church now has DSL with wireless). Of course, with my two kids running around after RE (religious education), my wife wouldn’t exactly be happy if I sneak out to the office to work on the podcast!


Editing Audio

I think it is important to have a short introduction and conclusion. I find that when I am listening to podcasts I don’t always have access to the file info, so without having an audible intro I have no idea what I’m listening to. For our podcast I read just about the same script every week, changing the date and the contents. For example:


Greetings and thank you for downloading the Unitarian Church of Marlborough and Hudson’s weekly podcast. Today’s recording was made at our service on Sunday, February 5th, 2006. Our introductory music is "The Tapestry," by William Blake and Audrey Snyder, performed by the UCMH choir. This recording features the Reverend David Boyer’s comments and interpretive readings from Dave Barry’s book, "Dave Barry Turns 50." Reverend Boyer’s sermon, "A Touch of Grey," follows the reading. More information about our church can be found on our website at UCMH.org. Thank you again, and enjoy.


At home I record the intro and conclusion using the iRiver’s built in mic. I then I grab the MP3 files off the iRiver (connects with a simple USB cable). I’m using a notebook computer running Windows XP Pro. I recently flashed the firmware to allow me to plug the iRiver into my linux machine. So, now I’m producing the PodCast on my Ubuntu Linux desktop.


I use Audacity to edit the files. There is a bit of a learning curve (especially if you’ve never done audio editing), but this free software can do just about everything. I edit down the long recording into just the short clips I want to keep. I maximize/normalize all tracks, then use each track’s volume adjustments to get the to sound ‘the same’ level. I’m looking for a way to automatically do this using the average volume level, but haven’t found this in Audacity yet (used to do this in Goldwave).


Audacity can export as MP3 using the Lame toolkit, but I personally do not export as MP3, because the export options aren’t specific enough for me. I export as a wav file. I follow a standard naming scheme – UCMHYYYYMMDD.wav.


The Lame tools can be used via the command line, and I also downloaded another package called WavTools. I use the WavTools wtrtc command to normalize the Wav file. Then I run Lame to convert from Wav to MP3. Here are the example commands:


wtrtc /N1.0 UCMHYYYYMMDD.wav UCMHYYYYMMDD_N.wav


(this creates a second version of the wav, but normalized)


lame.exe -h -s 44100 -m m --abr 56 --ta "UCMH.org" --tt "YYYYMMDD Service Title" --tl "Speaker's Name" --ty 2006 --add-id3v2 UCMHYYYYMMDD_N.wav UCMHYYYYMMDD.mp3


(this encodes the second/normalized wav file, I then delete the normalized version, I keep the original for burning to CD)


The lame options basically tell it to encode in mono, average bit rate of 56kbit, and also sets the ID3 tags. This results in a highly compressed but still very good sounding file. We run about 1/3 meg per minute (most mp3s out there run 2 megs per minute !). This results in our 30 minute files being about 10 megs – which equates to around 25 minute download on 56k, under 3 minutes on broadband. So, even on a modem you should be able to effectively “stream” the audio, as it downloads faster than it plays!


I wrapped these commands in a perl script to give me a menu when I need to encode.


Uploading/Hosting

I manually upload the mp3 file using FTP. I suggest FileZilla as a good and free FTP client. I now use gFTP on Linux. I have pretty high storage and bandwidth from my host – Hostek. They have both windows and linux servers, our site happens to be hosted on a linux box. I host a number of sites with them, so I’m set up as a re-seller. I get 2 gigs of storage and 50 gigs/month of bandwidth to share among all my sites, and I pay $25/month. I’ve also heard really good reviews about dreamhost – but haven’t had a chance to really look into them yet. I’ve moved our site to Dreamhost, they have amazing prices for the storage and bandwidth, linux only. Right now we are running about 415 downloads/month – which equates to 4 gigs of bandwidth. If we ever approach the bandwidth limit we can start thinking about using bittorrent to serve up the file. I haven’t looked into this much because apparently itunes doesn’t support it.


Content Publishing/XML Creation

Our site is built on a WordPress framework. WordPress is an open-source PHP and MySql powered blogging/content management application. It has RSS and podcasting support right out of the box, although I’ve added the WP-iPodCatter plugin to add itunes tags to the RSS (I’ve also done some major modifications to the system, and use my custom code-base for our chuch and also another site I’m the webmaster of). After uploading the file I simply write a post (usually with some pictures, an Amazon powered book link, maybe the text of the sermon) with a link to the mp3 file, and saving the post in our “podcast” category. The system automatically includes this new post as an entry in our podcast XML, and auto-adds all the podcast and itunes specific tags for the mp3 file. In addition, this also builds out the link on our site to download the file directly.


http://www.ucmh.org/category/podcast/ – html version of podcast category
http://www.ucmh.org/podcast/ – xml version of podcast category


Testing

Now that iTunes supports podcasts directly, it is the largest aggregator out there. I personally don’t use it for my podcasts, and I don’t have an iPod. I listen to podcasts on my Palm Zire 72 or my LG vx8100 cell phone (or our church’s iRiver), storing the files on a 512 meg mini-sd card. I use Juice as my podcast aggregator, so to test our podcast I just run Juice and see what I get!


Post Production

CDs
Besides providing the recordings of our services as PodCasts, I’ve started burning CD’s. My hope is to build up a library that congregants can borrow from. The sound quality is slightly better than the MP3s, and for those members without internet connections or slow connections, having the recording on CD is a real service. I take the main track and split it up, as a single 30 to 45 minute track is hard to use, but 10 to 15 3 to 4 minute tracks are easy. I also provide tracks of just the music, which has already been found to be useful for review during our choir rehearsals.

Listings
The majority of our PodCast subscribers are using iTunes, so by far that is the main place to get listed. Here are the other PodCast listing services I’ve found and asked to list our PodCast:
Yahoo! Podcasts
New England Podcasting
iTunes
Odeo
Podcast Pickle
Podcast Alley
Unitarian Universalist Podcast Directory
iPodder.org
GigaDial
podnova
digital podcast
Podcast.net
podfeed.net
TUUT: Talking UU Technology


Links

iRiver T30 at Amazon


Archos Jukebox Stereo Microphone at Amazon


NEW Ubuntu Linux


Audacity
http://audacity.sourceforge.net/


Lame
http://www-users.york.ac.uk/~raa110/audacity/lame.html


Wavtools
http://www20.cds.ne.jp/~yugen/prgw/wavtools/en/index.htm


FileZilla
http://filezilla.sourceforge.net/


Hostek
http://www.hostek.com


Dreamhost
http://www.dreamhost.com


WordPress
http://wordpress.org/


WP-iPodCatter: WordPress iTunes Podcast Category Plugin
http://garrickvanburen.com/wpipodcatter/


Juice
http://juicereceiver.sourceforge.net/index.php