I’ve written some notes about how to make and publish podcasts. Please leave a comment or email me if you have any suggestions or corrections.
The steps in making a podcast are:
- Create your audio file
- Edit your podcast
- Create the finished podcast audio file
- Put the audio file somewhere accessible on the internet
- Create a description of the podcast episode in RSS2.0 format
- Publish the description in a RSS2.0 feed for your podcast
- Validate your RSS feed
I’m assuming that you have already worked out what you want to say and probably written a script to help you keep on track while making the recording.
1. Create your audio file
The audio file can be recorded using any digital recording device. As well as digital audio recorders or a computer, many mobile phones and mp3 players also have the ability to record. But remember that you need to know how to get the recording onto your computer for editing and creating the final podcast. You can even use a digital video camera and just use the sound track.
It is much easier to improve the quality of recording by using a good recording setup than it is to try to fix it in an audio editor. So test the options you have available and see which produces the best ‘raw’ audio. Because a podcast is distributed as compressed audio, you want to have the clearest original sound, so that bandwidth isn’t wasted in compressing noise but in making your recording sound as good as possible. Changing the type and level of compression or repeated re-compressing can also introduce noise, so try to either use uncompressed recording if possible such as WAV or AIFF. If your recorder can only record in mp3 format, try to use the same settings as you want for the final podcast. As a general rule external microphones will produce a better quality recording, although the quality of the small microphones built into computers and phones are very good now. If you have alternative microphones, try them out and see which you like the sound of the best.
On a computer, the most sound editors (like Audacity or Apple Garageband) can be used for recording as well as editing. There are also specialised recording software just for making podcasts, for example Ubercaster on the Mac – see the resources section below for more information.
If you want to hear what the recording is sounding like while you are making it then you will need to use a headset, otherwise the monitored sound will interfere with the recording.
It is a good idea to record a few seconds of ‘silence’ at the start of the podcast – not true silence, but the noise in the environment and your recording system. This can then be used in an editor like Audacity as a ‘noise’ filter if you find your recording has too much background noise. However, noise filtering often introduces artefacts in the recording such as making the recording sound ‘bubbly’, so it is much better to make the recording in a quiet environment instead. Sometimes the noise is from the equipment you are using, and if this is causing a significant loss of quality you probably need to try a different setup.
When recording on computer, you can also set the microphone volume. It should be set to give a strong signal without ‘clipping’. ‘Clipping’ is where the recorded volume goes beyond the maximum that the recorder can manage and so some of the sound is lost. Set the level using what is likely to be the loudest part of the recording.
At this stage, experiment with creating a small sample of a final podcast using the equipment to make sure that you can bring the audio right through to the finished product without having transfer or conversion problems. Don’t wait until you have a full podcast recorded only to find it isn’t in a format / device you can use.
Audio Formats for Podcasting
Although many people now have broadband, I recommend that you always use the smallest size file that still gives the sound quality you decide you want to have. The same applies when making graphics and video files. This way people with lower speed internet connections may still be able to access your content, and it will be faster for everyone.
The most common format for audio podcasting is mp3, and this will be accessible to the widest audience. mp3 files can be recorded at a wide range of sampling rates and bit rates, but the most commonly used for podcasting is a sampling rate of 44.100kHz and a bit rate of 64kbps. Many mp3 players are unable to play more than 44.100kHz recordings, so if the sound recorder you have used produces more than this you will have to use an editor to reduce the bit rate. The bit rate controls the audio quality and file size. For mono spoken voices a bit rate of 48 kbps will still sound very good.
One way of deciding which format to use is to look at podcasts with similar content to what you are planing and see how they are encoded. On Mac OS X this is displayed by choosing "Get Info" with the podcast selected in Finder or iTunes. For example the ABC Radio In Conversation podcast (http://www.abc.net.au/rn/inconversation/) is encoded in mono, at a sample rate of 44.100kHz and bit rate of 64kbps. By recording in mono you only need one channel of sound instead of two which will make the recording much smaller for the same quality. So only use stereo if it is really necessary to the type of recording you are making. Generally spoken word recordings don’t gain a lot from being in stereo. This episode of In Conversation was 21 minutes long – so about 0.5Mb/minute.
If your target audience is known to have slow internet connections, you can try recording at the lower sample rate of 22.05kHz and the bit rate to 22kbps and see if the audio quality is still adequate and how much the file size is reduced.
2. Edit your Podcast
Use an audio editor on your computer like Audacity, Garageband or Ubercaster to edit your recording. You can cut out any bits that didn’t work the way you wanted, rearrange the order of the segments, and add in music clips or sound effects. If you are adding in other peoples audio, remember to make sure it doesn’t breach copyright. There are a lot of sources of free or low cost sound bites for adding to podcasts. Garageband includes a lot of free clips. The Creative Commons podcasting legal guide in the resources section below includes details of the copyright issues for podcasting and links to places to find podsafe content.
3. Create the final Podcast audio file
When you are happy with your edited podcast you will need to save or export it in MP3 format, making sure that the sampling and bit rates are set appropriately for your recording. This is most likely to be a sampling rate of 44.100kHz and bit rate of 64kbps. Depending on the software you are using, you may also need to choose the Bit Rate Mode and Channel Mode. I suggest that you set these to "Constant" and "Joint Stereo".
MP3 Export Settings panel in Audacity:
4. Put the audio file somewhere accessible on the internet
The mp3 file that you have created needs to be put on a webserver on the internet before you can publish it as a podcast. Depending on where you are hosting it you may upload it through a web form or use an FTP program to upload it.
There are a variety of ways to get hosting space. If you have an internet connection, it will probably include a small amount of webserver space as part of your service. There are some free file hosting services, but most of them don’t allow uploading mp3 files. There are free blog hosting services like blogger.com and wordpress.com that are a quick and easy way to setup a blog, but don’t allow the upload of .mp3 files as part of the basic service. By paying a fee, you can upgrade either the file hosting or blog hosting services to include mp3 files. The most comprehensive option is to get your own webserver and domain using a webhosting service. The cost for this will start at about $100/year. There are more details about the hosting options in the resources section below.
5. Create a description of the podcast episode in RSS2.0 format
If you are comfortable with editing HTML or XML files, you can just create the RSS2.0 format file in a text editor. Some blogging software has the ability to support the upload of mp3 files and setting the podcast tags so it can create the RSS feed. There are also editors designed just for creating RSS feeds.
The RSS2.0 specification is defined by the RSS Advisory Board – http://www.rssboard.org/rss-specification. The definition can also be extended or specific purposes and Apple has done this for iTunes.
A sample RSS2.0 file for a podcast (a single item extracted from the feed for Einstein a Go Go on RRR Melbourne http://rrrfm.libsyn.com/rss/Einstein%20A%20Go%20Go)
< ?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:cc="http://web.resource.org/cc/" xmlns:itunes="http://www.itunes.com/dtds/podcast-1.0.dtd" >
<title>RRR FM (Einstein A Go Go)</title>
<description>Podcasts from Australia's best known community radio station</description>
<copyright>Triple R Broadcasters Ltd.</copyright>
<webmaster>firstname.lastname@example.org (Liberated Syndication)</webmaster>
<generator>Liberated Syndication - libsyn.com</generator>
<lastbuilddate>Sat, 15 Mar 2008 04:00:00 GMT</lastbuilddate>
<itunes :keywords>Byte Into It, Breakfasters, Music Interviews, Dirty Deeds, Einstein A Go Go, Film Buffs Forecast, Healthtrip, Plonk, Radio Marinara, Radio Therapy, Run Like You Stole Something, The Architects, ALong For The Ride</itunes>
<itunes :author>Triple R</itunes>
<itunes :image href="http://libsyn.com/podcasts/rrrfm/images/rrr-logo.jpg" />
<title>RRR FM (Einstein A Go Go)</title>
<title>Einstein A Go Go Podcast: Mar 9th 2008</title>
<description>< ![CDATA[The genetics and biochemistry of night owls versus early risers. How an intelligent MRI machine can read your mind. The first ringed moon spotted by Cassini near Saturn. We heard from our guest, Amanda Gillon from the Department of Biochemistry at La Trobe University, about her work on cyclotides (circular proteins) and how they may lead to a new class of natural, more ecologically friendly insecticides. Plus the magnetosphere, and a strange link between long legs and galaxies. <br/><br />For more information visit <a href="http://">www.einsteinagogo.net<br /></a><br />]]></description>
<category>Einstein A Go Go</category>
<pubdate>Sun, 9 Mar 2008 15:00:00 GMT</pubdate>
<itunes :keywords>science communication</itunes>
<itunes :author>Einstein A Go Go</itunes>
<enclosure url="http://media.libsyn.com/media/rrrfm/Einstein-A-Go-Go-20080309.mp3" length="11436957" type="audio/mpeg"/>
The first section describes Channel information – the information that remains the same for all episodes of your podcast. Then each individual episode of your podcast is included as an “item” in the RSS file. The CDATA tag is used to allow you to include HTML formatting and links inside the RSS feed. This can make the outlook of the feed look better in some browsers, but makes it easy to include mistakes in your feed.
If you want to have your podcast listed on iTunes or displayed well in the iTunes client you will need to add the itunes specific tags to your RSS2.0 file. These extra tags are used to define the information iTunes displays. See the link to Apples podcasting specification in the resources section below.
For more details on ways to create the RSS file see the resources section below.
6. Publish the description in a RSS2.0 feed for your podcast
If you are using blogging software, the act of making a new post will automatically update the RSS feed. If you are updating the RSS file in a text editor, you need to put the updated RSS file onto your website, this is usually done as part of a blog, although a service like Feedburner will convert a link to the audio file in a webpage or blogpost into a podcast RSS feed.
7. Validate your RSS feed
Make sure that you use a feed validator to make sure that your RSS feed is correct. Errors may stop people from being able to get your podcast. One validator is http://feedvalidator.org/. It is also a good idea to subscribe to your own podcast and make sure it looks correct in iTunes or whichever other podcast client software you are using.
It is quite interesting to submit other peoples feeds to the feed validator. I was surprised at how few of the podcasts I listen to had a valid feed.
- Sample Rate
- The number of times the audio source is measured (sampled) per second.
- Bit Depth
- The number of bits used to record the audio – 16 bit audio can record 65,536 possible levels. 24 bit audio has 16,777,216 levels.
- Bit Rate
- How much data per second is required to transmit the file. In a non-compressed format, the bit rate is related directly to the sampling rate and bit depth. With a compressed format like mp3, the audio is compressed to match the bit rate you have selected. The lower the bit rate, the greater the compression and the lower the audio quality.
- Real Simple Sindication – an XML format for describing a podcast. http://www.rssboard.org/rss-specification.
- eXtensible Markup Language – a text format with tags that describe / define the contents of the file. The standards for XML are set by the W3C – http://www.w3.org/XML/.
feed the URL for the RSS file that is updated every time that a new podcast is added.
- iTunes Store
- the Podcast section of the Apple iTunes store is a a directory of podcasts, similar to the way that a search engine like Google is a directory of websites. The actual podcasts are not stored on the iTunes site, they are stored on the server used by the creator of the podcast.
- iTunes software
- a client program for receiving podcasts – the client directly looks at the RSS files on the various web servers for the podcasts that are subscribed to. So even if the iTunes site hasn’t been updated with the latest podcast, the iTunes client software will see it. (There are many other podcasting clients – see http://www.podcastingnews.com/topics/Podcast_Software.html for a list of them.)
- Joint Stereo
- uses various compression techniques to reduce the size of the file while preserving most of the stereo information. If you select Joint Stereo for a mono recording, it should not increase the size of the resulting file.
Creating a RSS feed
If you are using a WordPress blog then the PodPress website (http://www.podpress.org/) has useful resources and the PodPress plug in that makes publishing PodCasts in WordPress easy – it creates the RSS2.0 feed for you, including the iTunes tags and also tracks the number of people who listen to the podcast. If you have a website that doesn’t have an RSS feed, you can create a blogger.com or wordpress.com free blog account and then post short messages linking to your main site to generate the RSS feed.
There are many RSS editors available. Two commercial editors that include support for Podcast tags are one called Feed For All (available for Mac and Windows) which costs $39 USD. http://www.feedforall.com/ or a Mac only one called Feeder for $29USD http://reinventedsoftware.com/feeder/. Both have trial versions.
Especially if you are using a free hosting service, it is a good idea to use Feed Burner (http://www.feedburner.com/) to create your RSS2.0 feed. As well as making it easier to use the more advanced features of RSS, feedburner includes the ability to change the source of the feed – so if you decide to move to a different URL feedburner can redirect all your existing subscribers to the new URL. It also provides some good usage statistics and tracking. Feed Burner uses the URL for the feed created by your blogging software, or by you in a text editor and then reprocesses it into a valid RSS2.0 feed. You then update the links on your site to user the Feedburner RSS feed instead of the original one produced on your site.
Apple have a good summary page of how to make a podcast and list it in the iTunes store (http://www.apple.com/itunes/store/podcaststechspecs.html). They provide all the technical details of the supported audio and video formats and the way that you need to setup your RRS2.0 feed to work in with iTunes. They include a screenshot of an itunes item showing which tags match with the fields displayed in itunes.
The actual RSS 2.0 Specification – http://www.rssboard.org/rss-specification
A sample feed demonstrating the main features – feed://www.rssboard.org/files/rss-2.0-sample.xml
How to create RSS code that will work best across a wide range of viewers – RSS Best Practices Profile – http://www.rssboard.org/rss-profile-1
Validator for RSS/XML feeds – http://feedvalidator.org/
I did find one site that lets you upload your mp3 files and create podcasts. It automatically generates all the different types of feeds, including itunes. See http://www.podshow.com/
Unfortunately the free blog hosting sites like blogger.com and wordpress.com don’t allow you to upload mp3 files. They do let you link to podcasts that you have hosted elsewhere. Two that are recommended on the blogger site are Box.net which offers 1Gb storage with a 10Mb max file size for free and MoveDigital has a 30 day trial and is then $9.95 for a year. The limits for most free hosting means that they would only be good for experimenting.
If the prices for upgrading a file hosting service are monthly, then you should compare them to a full web hosting service as they are often a similar price with the same amount of disk space.
WordPress.com will enable mp3 uploads if you pay for extra space. The extra space is added to the base storage which is now 3Gb so and the minimum extra space is 5Gb which costs $20/year. This would give 8Gb of space for your podcasts and other hosted files. If you have an internet connection, you should also check whether your ISP provides some free hosting as part of your internet connection – this may be enough to experiment with podcasting.
Audacity (http://audacity.sourceforge.net/) is an excellent, open source audio editor for both Mac and Windows. To make .mp3 files with Audacity you need to also install the LAME MP3 library (http://lame.buanzo.com.ar/).
Tech Tip 108 – Recording a podcast with Audacity – http://www.geeks.com/techtips/2007/techtips-07Jan07.htm
On a MAC it is also possible to use GarageBand to record and edit your podcast.
A video tutorial is available – http://www.apple.com/ilife/tutorials/ and select the podcasting tutorial.
On a Mac there is software called Ubercaster (http://www.ubercaster.com) that is designed just for podcasting. It features a streamlined user interface focused on podcasting. It costs $79USD (or $95 in a bundle with Feeder http://reinventedsoftware.com/feeder/). The 30 day trial lets you export 5 minutes of an episode that you create, but also lets you export 5 full podcasts during that month if you have an internet connection. The trial version appends a short demo jingle to the end of each show you create. If you were planning on doing a lot of podcasting I’d recommend trying it out. Features like being able to save a template for your podcast, support for recording interviews through iChat, Skype, Gizmo or any other chat application (which includes letting the guest hear the podcast until the interview starts) and “Autopilot” – a window that shows your script, notes and resources organised in a timeline for the podcast.
David Kaye’s Blogarithms website has a clear explanation of what bit rate, bit rate mode and channel mode to use for mp3 files for podcasting http://www.blogarithms.com/index.php/mp3secrets/
Legal Issues & Sources for ‘podsafe’ content
CreativeCommons have a detailed overview of the legal issues of podcasting and links to podsafe content http://wiki.creativecommons.org/Podcasting_Legal_Guide
More podsafe music content – http://music.podshow.com/