I have been asked about my podcasting workflow. This article outlines how and what equipment I use to record the e-Learning Stuff Podcast. This is only one way in which to record a remote panel based podcast, and I am sure there are numerous other ways in which to do this. I have also changed how I have recorded over the two years I have been publishing the podcast due to changes in equipment and software.
Key lesson is that there is more to podcasting than just the technical stuff…
You can’t just expect people to be online when you want them to be and be able to converse on your chosen subject just like that. Not to say I haven’t done that, surprised a few people and asked them there and then to be in a recording.
What we do is plan a recording day or two in advance and check what time would be convenient for all. I generally allow 90 minutes and expect to get a 40 minute recording out of that, or a couple of 30 minute recordings.
We also plan what subject we are going to talk about, though sometimes we don’t have a plan and we meander… this makes editing a nightmare, but makes for a more natural (and therefore easier to listen to) podcast.
If we are planning what we are going to talk about, I now use Delicious to tag the stories with a tag such as #els078 so that other people can quickly find the articles and links. It also allows others to tag relevant stories they want to talk about.
For those who have never done a podcast before it can help if they prepare and have an understanding of how the podcast works. There are simple etiquette rules that need to be followed if the recording is going to be smooth and need minimal editing. Also it should ensure a better listening experience. I ask participants to be in a quiet room with minimal external noise. I once had someone who was in the server room (as that was the only place Skype was allowed) which meant that the recording didn’t work as there was this constant machine humming noise in the background. It also makes sense for them to test their headphones and microphones (and make sure they use headphones and a decent microphone). Though people are more often than not forgiving of poor video quality (look at the popularity of YouTube), they are less forgiving of poor audio quality.
My podcasts are either recorded from my computer or in the field.
In the field, for example at a conference, I will use the Edirol R-09HR audio recorder. This is a great little field recording device that gives me good quality and records the files onto a SD Card.
I record using uncompressed WAV so that subsequent editing and compression doesn’t impact on the quality of the recording.
On my computer (and I have the same setup now at home and at work) I use a variety of hardware and software to achieve the recording.
My preferred microphone is the Blue Snowball, due to the warmth it gives my recordings. Previously (and sometimes at work) I will use a Samson C01U microphone, however I have had issues with it as the pickup seems rather quiet and often a fair bit of post-processing is needed to manage this.
I bring the panel together using Skype, it’s free, easy to install and supports multiple participants.
I now use Ambrosia’s WireTap Studio to record the podcast. This allows me to record both the Skype input and the microphone input simultaneously as two different sources or tracks. This then enables me later to extract a track if required or boost the levels on one.
Previously I used Quicktime Pro to record the podcast. In order for Quicktime to record both the Skype and the Microphone I use another tool from Ambrosia called WireTap Anywhere. WireTap Anywhere allows me to combine multiple audio inputs into a single audio input. I can then set Quicktime Pro to record that single audio input. WireTap Anywhere can also be used if there are additional audio inputs, say there were two of us in the “studio” or someone was using another audio tool instead of Skype.
For the external participants I create a Skype conference. This allows multiple people to partake in the podcast recording, but the chat facility allows links to be shared as well as news stories. Participants also use the chat to inform me (as host) what they want to talk about next, or the fact they have to go so can we wrap the recording up!
In theory I could use Skype to bring someone in via the phone, but have not tried that. Once someone did use Skype on their iPhone, but we had to halt and start again due to problems. Where possible I ask people to be on a wired connection rather than a wireless connection when communicating via Skype.
I do spend a little time checking the levels of the participants and the location of their microphones to ensure that they are all at a constant level and they don’t constantly “breathe” into the microphone which can be off-putting to listeners. If I had the resources and the right kit, I would use a mixer and bring all the participants in individually so I could adjust the levels that way, however I don’t have an audio mixing board, neither physical or through software, so I am just playing or guessing that it sounds okay. If I was to go more professional than I would get in a proper audio mixing board. Bring in the delegates as separate channels, record those channels individually, before processing the final recording. I would also need to feed those individual audio feeds back into an external feed so that the participants could hear me and each other.
In the past, due to my poor broadband connection, we often had issues with Skype falling over and dropping the connection. Since I upgraded to BT Infinity I now rarely encounter that problem, though Skype outages have got in the way now and again.
Topics come and go, and often we start talking about one thing and end up talking about another. Theme shows seem to be popular and getting a panel to discuss something in particular seems to work well.
Topics usually come about after a chat on Twitter or face to face and we decide that would make a good podcast, so we arrange a time for a recording. Always happy to hear requests for shows on various topics.
So after starting Skype, I will bring the participants into a group chat. Usually rather than wait for a set time to record, I just hit record and record the entire conversation until we all hangup on Skype. As a result I often get some “out-takes” which often find themselves added to the end of the podcast during the final jingle.
Key rules for participants that I ask to follow are:
- Only one person speaks at a time. This ensures that the listener is not faced with multiple voices, something that is difficult in a face to face situation, but nigh on impossible to work out what is been said during a recording.
- Be aware that visual clues are “absent” and therefore not to keep rambling on because you aren’t been interrupted.
- Avoid jargon and explain terms, as though we as a panel may know what you mean, doesn’t mean the listener does.
During the show I and the panel will post relevant links into the Skype chat, and I often make notes using pen and paper that I will use later when writing up the podcast blog post.
As a host there are a few tactics I use to hopefully create a better show. I for example will bring people in who have not said much, ie “That’s a good point David, what do you think Ron?” I will also move the conversation on if I think we’ve run out of steam in that area. I try and keep the recording to under 30 minutes if I can and will wrap up the conversation, often asking the panel members for some final thoughts or comments on the topic in question.
Having had our “conversation” I will then (if not done already) record both the introduction and the end of the podcast. In the introduction I introduce the topic and the panel. At the end I thank the panel and ask where they can be found online.
Trying to get a good show title is always challenging, as you want something that either describes the show or catches the eye of people flirting through their RSS feeds.
After pressing the stop button on WireTap Studio I am left with a WAV file within the application.
I can then adjust the levels of the Skype conversation and my own microphone recording before saving the recordings. This is non-destructive so I can go in and edit again if I need to.
I then export the recording out as a WAV file. A typical 30 minute show at this point is about 400MB.
If I am not happy with the levels, and generally this may be a problem with the Skype part of the recording, someone was speaking louder than the rest of the panel, I want to normalise the levels so that there is a constant volume across the different voices. If each voice was on a separate channel than this would be quite easy to do, but generally they are not. I have had similar issues with the Edirol R09H. As a result I use an application called Levelator. I use Levelator as though it would be possible to manually adjust the audio levels, this would take a lot of time, Levelator “improves” the file faster and easier than I have the time for.
I then use Apple’s Garageband to create the final podcast edit. Garageband is a multi-track audio editing application.
Having imported the audio file into Garageband I carry out the following tasks.
Firstly trimming the podcast, removing extraneous content, though I may keep short humour clips for the end music. If I have made multiple podcast recordings during one session I will need to cut out those.
Secondly removing pieces which either went wrong or don’t fit or we went off on a technical tangent or because something was said that shouldn’t have been said. Generally I don’t need to cut things out as people are careful about what they are saying, but at least once I got an e-mail from someone saying I shouldn’t have said that about this or a person. So I will cut that out if asked. It’s more difficult to do after publication, so hence preparation and care during recording is a preferred option.
I also need to ensure that the introduction is at the beginning and the end is at the end
Sometime I need to remove silent bits, this can be because the panel were thinking or because of Skype delays.
I also add the intro music and the endpiece. These are jingles which are included as part of Garageband. I keep meaning to write some original pieces, but haven’t got around to it.
I also add metadata to the recording so that it identifies the title, who was on it and some info on what it was about. I also add a “podcast track.
Once I have edited the podcast in Garageband I export as an mp3 file. I use the 128Kbps setting as I have found using the lower 64Kbps can compromise the recording and reduce the quality.
256Kbps is nice, but file sizes can be large and not everyone has fast broadband. Using 128Kbps results in the 30 minute recording been about 30MB, which is a 90% reduction on the 300MB of the original WAV recording file.
I save and retain both the Garageband file, the original WAV file and if undertaken the “levelated” WAV file. As a result each podcast takes up about 1GB of backup storage.
One problem with exporting as mp3 is that track artwork isn’t exported with the file, it is if you use AAC and that certainly is an option. Garageband can not export an enhanced mp3 file, though it can export an enhanced AAC file.
My solution was one I found on the web. I open the mp3 exported file in iTunes. I can then drag a PNG file I created onto the album artwork and add artwork to the mp3 file.
The reason for this is that when someone listens to the mp3 file on their computer or their iPod the artwork will be visible. It helps with brand awareness and reminds them what they are listening to. I have considered creating individual artwork for each episode, but that is something else I could do, but to save time I don’t.
So after all the recording and editing I am left with an enhanced mp3 file that need to be distributed. At this point it is merely an audio recording and not a podcast, in order for it to be a podcast it needs more work.
Uploading the podcast
There are various ways and methods to publish a podcast. Garageband allows you to do this with a MobileMe account for example, using the Feeder software I discuss later also allows you to do it too.
The first thing to consider is where are you going to upload the podcast. This is something that needs some consideration, as if hosted on your home computer (possible) it could impact on your browsing experience as multiple people download it from your home server (computer) and if you have a bandwidth cap, it could impact on that. You could host it on your website, but again there may be bandwidth caps on traffic or in some cases the service provider doesn’t allow you to host audio files on your web server. There are commercial providers such as CacheFly, Amazon CloudFront amongst many that provide such services.
I currently upload the podcast to my WordPress.com blog account. This is not a free service, I pay $20 (about £12) per year for the 5GB space upgrade which allows me to upload mp3 and aac files to the blog. I can then link to the audio files from my self-hosted WordPress blog where e-Learning Stuff now resides. The advantage to me is that those people downloading the podcast won’t impact on the web server hosting the blog and the cost of storage space on WordPress.com is much lower than the equivalent storage and bandwidth costs on my web server.
I will write a blog post describing the podcast episode, this has a description, who is in it and what episode it is.
I use to use the Audio player plugin for WordPress so that I can embed the audio recording into the blog post direct. This means that you can listen to the podcast direct on the blog. However it uses a Flash player and as a result wouldn’t work on a mobile device such as the iPad that doesn’t support Flash. After some research I found another WordPress Plugin called Degradable HTML5 audio and video Plugin.
WordPress plugin for html 5 audio and video allows you to embed media for native playback in supported browsers and offers a graceful degradation for unsupported browsers.
This works on browsers on a computer (Mac or PC) and on mobile devices such as the iPad, iPhone and Android devices. What is nice for the listener is that they can start listening straight away without needing to download or transfer. More difficult to get stats on this, but great for the end user. I also provide a direct download link, so that people who want to use it on a particular device can just download and transfer it. After that there is a link to subscribe direct to the podcast in iTunes and later I will describe how that was done. I finally finish the post with notes from the show itself. These links are taken from the Skype chat and from handwritten notes I make as we record the show, though I sometimes make the notes on the iPad now. More often than not I will add a photograph to the blog entry to add a bit of colour (though it has been usually a black and white photograph). These photographs are either from my own collection, my photos on Flickr or found through a search on Flickr.
The post is then published, I can schedule posts in advance, and will do this if needed or I know I am out. I usually publish the show on a Sunday, this then gives people time to find and download the show ready for their Monday morning commute. Though talking to people some listen to it at work, whilst out walking, at the gym or on the train.
Podcasting the podcast
At this point I could say job done, but the real value of a regular podcast is that listeners can subscribe to the podcast and it then is “pushed” to their computer or device so when they want to listen to the show it is already there.
The key to this is producing an RSS feed that can then be subscribed to by the listener. It’s not necessary for the podcast to be in the iTunes store for the end user to subscribe to the podcast; they can subscribe to the podcast in iTunes (or a browser) just using the RSS feed. Once subscribed the end user can either listen to the podcast on their computer, sync it to their iPod, or manually transfer the mp3 file to another device.
So how do you create the RSS feed?
Using WordPress in itself creates an RSS feed that can be used for subscriptions. The problem I had with it, was I wasn’t just using the blog for the podcast, I was also posting other blog articles. The default WordPress RSS feed contains the last ten items, so more often than not the WordPress RSS feed will only contain the last one or two podcasts and not the last ten or twenty that I wanted to appear. This isn’t so much a problem for regular listeners, but someone who had just discovered the RSS feed wouldn’t be able to discover previous episodes without a bit of work…
Another problem with the WordPress RSS feed is that it is not “good enough” for the iTunes Store so in order to make work, and ensure that as many fields as possible are filled in I use to use Feedburner to take the WordPress RSS feed, augment it with iTunes Store friendly information and spit out an iTunes Store friendly RSS feed. Feedburner would also just take just the items with audio attached to them. However even after doing this “trick” I still had issues with the feed, in the main as it would also add non-podcast recordings to the feed.
I eventually came around to the conclusion that the only way that this was going to work, was if I created the RSS feed “manually” and used that instead of relying on the automated feed from WordPress. At that point I was resigned to using Notepad to write my own RSS XML file when I found out about Feeder. It is now a very simple process to create the different podcast entries that populated the feed.
With Feeder, you create an RSS feed and then can add entries to that feed before publishing the feed. What I do is create a new item.
Add the title from the podcast. I insert the article link from the blog, the posting that I wrote previously about the podcasr recording. There are iTunes fields that need to be filled in, including who’s in it and the duration. I add the direct link to the mp3 recording (from the WordPress.com server) and add a description. I use an edited version of HTML source from my blog post as the description and then I copy this over to the iTunes summary.
This is then saved and published to a web server.
The configuration also allows you to add information to the different iTunes fields required if you are going to submit your podcast to iTunes.
For legacy reasons I do import that RSS feed into Feedburner, but you don’t need to, I only do this because I already had people subscribing to that feed and I didn’t want to lose them as regular listeners.
I have configured the RSS feed to be the last twenty episodes, however with Feeder it was also possible (following requests) to create a full feed of all my podcasts.
This RSS feed can be posted or linked to and people can subscribe to it, though of course what most people want to do is to have their podcast within the iTunes Store itself.
Submitting to iTunes
The main reason for been within iTunes is for discovery, people looking at other podcasts may “discover” yours. You also have a “page” within the store and a link that you can pass to others, making it easier for people to know about the podcast.
The process is quite simple, but I initially waited until I had a few episodes done before submitting the podcast to iTunes. This would show them that I had some credibility and wasn’t just testing the system. They can be quite strict on this, so I do recommend recording a few episodes before going through the submission process.
To do this you need go to iTunes Store -> Podcasts -> Submit a Podcast (on the right hand side) or click this link.
You will need an iTunes account and you should check your RSS feed before submitting it.
It took only two days for the podcast to be accepted. As a result I now had a page in the iTunes store and an URL I could hand out or link to for the podcast.
So that’s how I do my podcasts. Not a simple process, but it works for me.