Podcasting How To

1. Introduction: How I podcast

A word of warning in the beginnin: This "how-to" is not thought to be a thorough tutorial about podcasting, all it's technical possibilities and tools. Also I am no "pro" podcaster. I just record lectures occasionally, made a podcast series for the Austrian Computer Society in summer 2007 and maintain a podcast for our software-engineering lecture.

Having said that, making podcasts was a learning process, and I also made some significant mistakes. As I was asked in the last months several times by students and other colleagues how I would do this and that task, I just decided to write down my procedures, maybe it is helpful for someone out there.

As I am not a pro, I am of course interested in suggestions to improve the quality of my podcasts, so if a reader has some suggestions to make or some violent dissent, please do not hesitate to tell me!

Another general comment: I am working with Macs; I also believe, that the Mac is the best platform for recording and processing Podcasts. However, there are also some tools for Windows out there. In this short text I only talk about the Mac tools I use, as far as Software is concerned.

top

2. Podcasting?

I think it is not necessary to describe in general terms what podcasting is, as your intention is to produce a podcast, however, I want to conclude what type of podcasts I am doing and what this howto is describing

• Pure audio podcasts, recorded live or with mobile equipment
• Enhanced podcasts, i.e. Podcasts that include pictures, figures, URLs and chapter marks.
• Recording from lectures and presentations

I do sometimes record interviews and group discussions, but I do this with single microphones and technically low-key, i.e. not involving a mixer and stuff. Sure, there is improvement there. Now let's dig deeper:

top

3. Recording 'in the Studio'

Well, "Studio" is quite a word, actually I am recording in my office at university or at home. This is sure a problem, because you "hear" the room, depending on the microphone setup. I use a Samson CO3U for that situations. This is quite a nice mic, has the advantage that the microphone characteristic can be adjusted. I use the Super Cardiod pattern when I record my voice alone, e.g. for Screencasts, the Figure 8, when discussing with just one other person, sitting opposite to each other and the Omni pattern when I record a group of people.

The first and most important thing is to adjust the record level properly. My first recordings used a too low level, fearing to clip, but when the level is too low, the signal/noise ratio is bad and the recording volume after all to low. This is particularly bad for people listening with iPods and stuff. Also the setup is important. People should have approximately the same distance to the microphone and all noise (knocking on the table...) around the microphone is to be avoided.

For recording I typically used Audacity (or Garage Band, but I generally prefer Audacity for the initial recording.). Audacity is an Open Source audio software, that is available for Mac, Linux and Windows. Take care to be sure to have selected the correct microphone! In my first attempts, I actually used the built in microphone instead of the external USB, because I did not use the setup accordingly. This is particularly true for GarageBand! I also use Audacity to do some audio processing steps (e.g. Noise Removal).

However, recently I bought WireTap Studio and this tool is really great. Not only is it possible to record from various sources (see below), but it also has a great audio editor, that allows to easily find a specific place in a long recording by a nice preview function. Additionally cutting and cropping is perfect: it is non-destructible, i.e. you can modify the edits later on, also the fade-in and outs. Great for recording, cutting and initial recording steps.

GarageBand on the other hand has an advantage when enhanced podcasts are recorded: what I recently noticed is, that already during the recording with GarageBand the "P" key can be used to set chapter marks. This comes in very handy for post-processing.

top

4. Recording of Skype Interviews

As mentioned already in the previous section, I meanwhile use WireTap Studio for recording. WireTap is also particularly good for recording Skype sessions (interviews.) Either use one recording channel set to Mac audio (not recommended) or better: set one recording channel to Skype (or any other VoIP tool) and the other channel to microphone.

Why is this "two channel" approach better? By several reasons: first "noise" from other applications like calendar reminders, email notifications and the like are not directly recorded. But even more important: in the wave-editor it is possible to set the volume of the two tracks individually. With a little trick you can also export both separately, then import both in two Garage Band tracks and e.g. apply different filters or arrange them in a stereo setup.

top

5. Recording of Lectures and Presentations

Recording lectures and presentations is more tricky, as typically also slides are involved. ProfCast comes in handy here: ProfCast is a nice and easy to use application. You take your presentation file, drop it into Profcast and start the recording. Profcast then starts the presentation software and starts recording your voice (check that the correct microphone is selected). Additionally Profcast creates an enhanced podcast by making chapter marks with every slide-change and also adds the slides as bitmap to the podcast.

Microphone-wise I suggest to use a Microphone that is attached to you, like clipped to youre clothes or something to get a proper recording result.

I typically use the Export to GarageBand function. The "raw" Profcast recording is hardly good enough to be used directly. Often Profcast gets chapter titles wrong, and also enhancements in the audio are usually required.

Another tip here: The GarageBand file is actually a package, and you can open that package. After pushing the ProfCast recording to GarageBand, the audio file is actually stored as aiff file in the package. This aiff file can be manipulated as described below with Audacity and Levelator, as long as two things are fulfilled: (1) The filename must not change (2) Do not cut the file in Audacity; just do filter manipulations, cutting, cropping and stuff has to be done in GarageBand afterwards.

ProfCast also offers other features like a podcast manager. I n my opinion it does not do the job right. How I solve the publishing issues, see below.

top

6. Mobile Recording

For mobile recording I use a AKG C1000S Microphone and a mobile recorder, the M-Audio MicroTrack II. However, there are a lot of different mobile devices that can be used. Again I think, that a very important issue is to keep the signal level high. Try out all options at home before you go "into the field".

The MicroTrack allows to record to WAV and MP3, I use MP3, because for spoken language in a podcast, the quality is way good enough and it saves space on the card. Also use the monitoring features of the device while recording. I once tried to operate the MicroTrack with the power pluged in and had terrible noise coming from the power adapter. If I would not have noticed that in time...

top

7. Post-Processing: Audio

Now, what are my usual "post-processing" steps directly after the recording:

• Reduce noise. This is a quite nice feature in Audacity. The noise removal filter is actually an enhanced noise gate. You need to have some seconds of "plain noise". So before or after the actual recording, record some seconds of noise. Then select the noise area, select the filter, make the noise profile, then select the whole recording and remove the noise. However, keep the parameters low; I usually do under 10dB. The stronger the filter is working the more "artifacts" it produces. On low levels it works very fine
• Normalise the signal, i.e. use the full dynamic range of the digital signal. This is important to keep the signal loud, particularly for mobile devices like iPods
• Cut the raw material, also check the Envelope tool, that can be very helpful to adjust the dynamic at very problematic places in the recording.
• Then I save the recording as aiff file.
• The next step is very important for voice podcasts (interviews, discussions...): Use Levelator: This is a very powerful, easy to use and free tool. It is kind of an enhanced compressor. It analyses the audio piece and levels the signal. E.g., if two people discuss, and one person is speaking in a lower voice than the other, Levelator levels the signal. Cool tool!

In Garage Band I do some final steps like adding Jingles and putting together interviews and the like. Occasionally I also use GarageBand filters like the compressor. When there is an "S" problem, i.e. the spoken S sounds very sharp I use the Equaliser in the "Reduce S" setting but I reduce the intensity as the default intensity is too strong in my opinion.

However a general hint here: do not rely only on monitoring the sound quality on one single output device. Try at least the speakers of a notebook, and two different headphones (probably also regular iPod phones as they are often used). E.g. the "sharp S" is often much stronger in headphones (earplugs) than in the notebook speakers, that can be misleading!

The next steps are done in Garage Band: I create a new podcast project, create a track and drop the aiff file into this track. In Garage Band additional operations can be done easily, like adding music, putting together different recordings, or creating an enhanced podcast as described below. Set the desired output format the GarageBand preferences (!) and write the podcast to a file. That's it for a simple audio podcast.

top

8. Post-Processing: Enhanced Podcasts

Having the finished project in GarageBand I typically produce an enhanced podcast: I add chapter marks with GarageBand, pictures or photos and URLs when appropriate. Set the output settings in GarageBand according to the desired output quality. GarageBand then generates an m4a file. This m4a file can be played with a wide range of players on all major platforms including iTunes, Quicktime, Miro, VLC, iPods and so on.

top

9. Hosting of the Audio Files

After having created the m4a or mp3 or mov (Screencast) file, I name it properly like "yyyymmdd-name.m4a" someting like that and upload it to a webserver using FTP, SCP or RSYNC. If you have no according infrastructure you can also check out services like Podster.de.

At that point we have audio files on a webserver but not yet a podcast, so let's do the final steps:

top

10. Publishing: Feed and Shownotes

The important point about a podcast is the podcast/RSS feed. This is an XML-based format that describes the podcast episodes (Title, Description, ...) and where the audio/video file can be found. This feed is used by iTunes and other Podcatcher to stay up-to-date. There are several ways to create this feed, I have done all of them ;-)

• Manual: yes, done that. Take a text-editor, get a format description e.g. the pretty good one from Apple and write it. However doing this is awkward and error-prone.
• Use the podcast Manager from ProfCast. Don't like that. Confuses me more than it helps
• Use Blogger.com or a similar blogging tool!

Actually for my purposes the last option is the most elegant and also fastest one. Yes, Blogger creates podcast Feeds! What you have to do is quite simple:

• Create a Blog on Blogger.com.
• For each episode create a Blog article. Write the title, a short description. This also has the advantage that you also have done the show-notes done that way.
• Upload the m4a... to your favorite webserver, and copy the URL to this file into your system clipboard.
• Add the URL as "enclosure" to your blog article (under the title of the blog entry)
• Press the publish button

Having done that, Blogger creates the blog article and the RSS feed of your Blog is immediately also a podcast compatible feed! That is actually really simple.

top

11. Enhancing the Feed

We are more or less done with the previous step. However, I recommend another web-service: Feedburner. Feedburner works perfectly together with Blogger (check the tutorial at the feedburner site).

Feedburner is actually doing this: It takes the "original" RSS feed from e.g. Blogger (in that case) and re-publishes that feed under a different URL like:

http://feeds.feedburner.com/best-practice-software-engineering

This is the Feed URL of one of my Blogs. It integrates well with Blogger, so if someone subscribes to your Blog feed at Blogger it can be redirected automatically to the Feedburner feed.

Now why would you want that redirection? Actually by three reasons:

1. Feedburner provides statistics on how many people have subscribed to your feeds; how many visitors you have on the blog and so on.
2. It can manipulate and enrich the feed, like by adding iTunes-metadata for you. Check out all Feedburner options.
3. Feedburner creates a feed that adapts to the client: it produces a page that is compatible with Browsers, iTunes and other Podcatcher.

top

12. Promotion and Last Words

O.k. not much here: you might consider putting the Blogger blog to the Blogger lists, add it to Technorati, Podster.de, the iTunes directory and the like.

So, I hope this very subjective how-to was useful to me; Write me if you have comments. And produce good podcasts!!

top