You’ve got all your shiny new gear ready to go, but how do you turn all that steel and cabling into podcast goodness? Our 9 expert EM podcaster (@FOAMpodcast, @srrezaie, @TheSGEM, @stemlyns, @embasic, @Core_EM, @EM_Educator, @EMtogether, @EMCases) plus a 10th new addition, Dr. Rob Orman (@emergencypdx) of ER Cast and EM:RAP fame, share their system for getting from idea to published podcast. Lets take a peek behind the acoustic foam board at podcasting workflows.
We scour the FOAM universe for recent posts that are interesting/cutting edge. We discuss options and consider which posts we could easily relate to a good core EM content topic that doesn’t get as much play in FOAM. We then use Google Docs to write the script (about 90% of the show is scripted ahead of time). 30 minutes of recorded audio generally yields about 20 minutes of content. One of us (JF) edits the audio and the other (LW) creates the blog post and tables/images.
We don’t mind the idea of repeating ourselves, by the way. We haven’t repeated any big topics so far but we are 100% ok with the idea of doing a future show on a topic we’ve done before, as spaced repetition is a crucial part of learning.
We typically have a theme (i.e. trauma, resuscitation, prehospital, respiratory, etc…) every month and find two recently published studies whether good or bad and do a critical appraisal of them.
A recent publication is picked by SGEMers using the Hot or Not feature on SGEM website. Listeners also contact me directly with suggestions of articles to review. Sometimes the person suggesting the article is asked to be the guest skeptic. Other times, I will contact an appropriate expert in the field to be the guest skeptic (Critical Care, EM Pharmacist, RN, EMS, Physiotherapist, EM, or other medical/surgical specialist). They complete the SGEM critical appraisal work sheet. I do an evidence-based medicine over read and create show notes. We set a recording date. The recording typically takes ~45 minutes. I try to edit the show immediately after to 20-30 minute length while it is still fresh in my mind. There is also a ~2,000 word blog that accompanies each SGEM episode with hyperlinks out to further information. The completed product is then uploaded to the website and iTunes every Sunday.
We do a special SGEM episode once a month called Hot Off the Press (#SGEMHOP). These are reviews of articles in AEM or CJEM that have just been published within the last month. An author from the study is invited on the program to discuss the strengths, weakness, and limitations of their research. The SGEM audience can then interact with the author via social media.
Ideas happen randomly: clinical quagmire on a shift, a question may pop into my head while on a bike ride, many ideas come from listeners. If I’m not near my desktop or laptop, I’ll record the idea into the iPhone voice memo app. When I get back to the computer, I start a Google doc. Google drive and Google doc are the hub where I keep all running podcast ideas. Sometimes a podcast outline will come about in minutes; there are also times that it can take weeks of research and revisions. If the show is going to be an interview with an expert, I share the doc with them so we can work on it together (and they have an idea of the main topics).
I have an idea, think about it for a while, and ask myself, “what problem can I solve for my listeners?” or “what do people really need to know to put this to use in the ED?” I am always thinking about what the listener experience is like, and what they need from me. I jot a few questions or main ideas to myself, and mull it over some more. When I have a more formed idea, I do literature searches to be sure I am up-to-date (and can offer references to my listeners). Then, I turn on the mic and drop some mad beats.
We have a few “series” that we are developing like the “Induction” (for doctors and others new to EM) and “Wellbeing” podcasts (with Liz Crowe) but otherwise tend to talk about things that interest us. We figure if we are interested in them there is a good chance the listener will be too. We’ll talk about a topic, maybe over email, and think about our three learning points and how we can them expand on that. We are always working to make EM accessible, offering listeners alternative and/or structured approaches to the speciality. We maintain a long list of important topics that we need to get round to at some point, but there is no harm in doing topical subjects such as conference reports that may crop up at various points along the way.
I will admit that my work flow is very haphazard and this is something I struggle with. I publish episodes as I have time available so this means a very irregular podcasting schedule. Sometimes I will publish two or three episodes in a month but sometimes it will be 3 months in between episodes. For anyone who is starting out podcasting, my biggest piece of advice is to not publish your first episode until you have a few “in the can” ready to go. This will allow you to stay ahead by at least a few episodes and publish on a regular basis. When I first started in July 2011, I prepared three episodes for release all at once because I wanted to publish them as quickly as possible before the new interns got much further along in the year. What I should have done is rolled out those episodes slowly over weeks to months so I always had new content ready to go.
For my podcast, I type out a script of every word that I say. In the beginning I tried to do it with just an outline but that didn’t work to well. The script writing is the longest part of the process- for a 20 minute episode I may write 8 to 10 pages single spaced so that can take 7 or 8 hours in total to accomplish. Then its just finding time that I have available to record the podcast which takes about 1-2 hours. Editing the episode takes about another hour and posting it to the website and publishing it takes about another hour as well. All told, one episode may take 15-20 hours of work to finish. My wife is a doctor as well (she is starting a peds critical care fellowship this summer) so to make sure we have time to spend together, I make sure to work around our schedules so I’m not sacrificing family time to produce the podcast.
Some people put a lot of effort into finding just the right space to produce a podcast by recording in a closet or even setting up sheets around their workspace. I have never found that necessary but I will make sure to minimize background noises. My “office” (aka our spare bedroom) is right outside our porch where we have a wind chime so I make sure to disable that and close the dog door so you don’t hear our two dogs going in and out of the house. Some people will turn off their air conditioner while recording but in Texas in the summer that isn’t really possible. To overcome this, I use an editing tool in Audacity to eliminate background noise.
I work on a number of podcasts (EM:RAP, EM Abstracts, REBEL EM and Core EM) and the workflow is a bit different for each one. Since I’m primarily responsible for Core EM, I’ll discuss my workflow for that one.
Each week, my cohost for the podcast and I will pick a topic from our conference to highlight on the podcast (see below for content). One of us will write up a script and then we’ll kick it back and forth over the week to edit. Early on, I find it helps to have fairly detailed scripts but as you become more comfortable with the process, the scripts can be come more and more bare bones.
Once the script is complete, we pick a time to record. Generally, we’ll record via Skype since my office has terrible acoustics. I’ll record both ends on Skype recorder and each of us will also have a local recording with sound studio. If the Skype recorder call had good quality, I’ll use that one and if not, I’ll merge the individual recordings and work off of that as the master. While we’re recording, I’ll mark any errors we make (explitives mistakenly dropped, hiccups, errors we rerecorded to fix etc) on the script with a time mark so that I can easily find them when I’m editing.
After the master is set, I’ll save it as a new file (one of the keys is never to delete or save over anything because you never know when you’ll need to go back and start over) open in my recording program and start editing.
After editing is done, I’ll use a program to “level” the sound so that the voices are relatively equal in volume. Two programs that do this are Levelator (free) and Auphonic (not free). Once leveled out, I add the intro/outro music overlay (I find this easiest to do in Audacity) and record as a completed file. I send the completed file over to my cohost to review and then load it up to the site (via Blubrry). Add in show notes and featured image and the podcast is ready to go. The hosting site (in my case Blubrry) makes sure that once published, the podcast makes it to iTunes.
As soon as the podcast is completed and scheduled, I’ll write an auto-tweet on Hootsuite to be released the same day.
While this is my current workflow, newer software is always emerging. There are some new on-line services that captures voices on individual tracks and automatically splices them together. These products would make lots of aspects of recording easier. To date, I haven’t found one that I like but I keep looking…
A good podcast starts with a good idea. Recently, a good friend and colleague of mine, Will Sanderson (faculty at The University of Kentucky) and I have been working on our latest podcast creation, Wildcast EM. In addition to Will, the brains behind this podcast include Jon Bronner and Chris Doty. Many of our good ideas (at least I think they are good) have been generated during discussions in coffee shops and pubs in Lexington, Kentucky. It helps to find an environment that helps spark your creativity. In general, we start with an idea and then sit and record some “practice sessions” in my home studio. Some sessions work and some don’t. In fact, a lot of our ideas fall flat. Podcasting requires a lot of practice to get the chemistry right. As far as scripting, we really don’t utilize that strategy at all. Maybe we should! Some podcasters write elaborate scripts that they use when the record. The problem with this is that the end product can sound rather robotic. Ever listened to a podcast that sounded scripted? I bet you have. Having a script is fine as long as you sound natural.
In general, I loosely storyboard a podcast like I do a presentation. I look at the idea behind the podcast then I create a flow to it. I look at things like the introduction, the body of the podcast, and how to close. I also look at how to integrate music into the podcast.
The general outline of a podcast is usually:
- Intro (maybe with some music)
- Body of the podcast (solo podcast or group)
Once you get more and more into podcasting you will want to break out of this mold and be different. I can certainly share some of my ideas in a future post or if you contact me.
For the Main Episodes:
Week 1: Research the topic using Rosen’s, Tintanalli, FOAMsearch, Trip, primary research papers, discussions with the guest experts and a variety of books in my library.
I believe in the educational principals of spaced receptive learning and multimodal learning and especially the notion that learning from multiple perspectives, opinions and approaches to common medical problems helps us become better clinicians. So I do not stray away from covering a topic that has been covered on other podcasts or blogs (e.g. Delayed Sequence Intubation). That being said, most of the less common topics come about from seeing a great talk (e.g. Backboard & Collar Nightmares), from previous discussions on EM Cases episodes (e.g. IV iron for Anemia), reading an interesting paper (e.g. Obesity Emergency Medicine Management), having local experts on the topic (e.g. Geriatric Emergency Medicine ) or from a formal needs assessment (e.g. all the pediatric topics on EM Cases). I’ve been thinking about a more formal approach, but that’s just not as interesting or fun.
Week 2: Develop a written outline that includes cases, key questions I might ask on the podcast, along with notes that provide a springboard for discussion based on Week 1’s research, and key references. Sometimes I use storyboarding to generate and organize the content; other times the questions flow on their own. Once I’ve got all the questions, notes and references done I write up an introduction, which is vitally important to engage listeners.
Record the podcast: this usually takes approximately 3 hours including set up, pre-recording chat, recording with some 2nd takes on occasion, breaks and thank you’s. Sometimes the flow is good and we’re done in an hour. Other times there are lots of stops and starts to get things right. And still other times the discussion so intriguing (to me at least) that I let it go for a long time. The podcasts that I record with Walter Himmel tend to take about 5 hours as he has so much great stuff to say. Then I edit it down to a 1 hour episode or two 1 hour episodes. I usually instruct the guests to speak in a conversational tone but a bit louder and more energized than they might in a normal conversation. I encourage them to use storytelling and examples to bolster their points as these methods are memorable for the listener.
Week 3: Editing and post-production, in my opinion, is the most important part of producing a podcast outside of the content itself. I spend a lot of time with this. I have some help doing some of the basic editing by medical students. There are several principals that I try to adhere by:
- Approximately every 10 minutes the listener needs some sort of mental break, whether it be a music jingle, a different voice (or my voice adding in content that sounds outside of the interview with the guests) or a comedic interlude
- Again, spaced repetitive learning with reviews of the material interspersed in the podcast
- Some ‘uhms’ and ‘ahhs’ are okay to leave in, but if there are many, some need to be edited out; the cadence of the speakers needs to sound as natural as possible when splicing and dicing
Steps in Editing/Post-production:
- Run the raw recording through Auphonic (which Scott Weingart kindly introduced me to) to get rid of any background noise and hum.
- Dump the audio into a GarageBand template that includes the EM Cases introductory theme music.
- On the first pass listen I edit out gross chunks of audio, ‘uhmms’ and ‘ahhs’ if they haven’t been editing already by the medical student, and any re-organizing of content; all the while taking notes for the written summary that will come later.
- On the 2nd pass I do finer editing, adding reviews and comments, stingers, jingles and sound effects.
- I listen back to the whole thing while jogging, in my car and on my home stereo for any additional content editing that might need to be done and to make sure is sounds good in all 3 settings.
- Run the recording through Auphonic again to get the noise reduction and loudness correct.
- Use ID3 tag editor to tag the file.
Week 4: Posting the podcast involves loading the podcast onto the website using WPfilebase, choosing a non-copywrited image that evokes emotion or intrigue to go along with the post and reworking the written summary of the podcast completed by an EM Cases Team member into a blog post. I test all the links and then release the podcast via Blubrry Powerpress to iTunes and separately onto SoundCloud.
For the Journal Jam podcasts:
- Justin Morgenstern (@First10EM) interviews a researcher and then Teresa Chan (@TChanMD) and I listen to the interview and weave in our commentary at the EM Cases Studio.
Once I’ve released the podcast, I post an announcement on Twitter, FaceBook, Google Plus and Linked In. I also send out an email blast via MailChimp to the 5000 or so EM Cases subscribers. I have found that partnering with other online educational resources and trading links, maximizing Search Engine Optimization, offering multimodal learning that is based on the podcast such as eBooks e.g. EM Cases Digest and The EM Cases Course, partnering with prominent conferences (e.g. North York General Hospital’s Emergency Medicine Update), giving talks and workshops at conferences, have all directly or indirectly helped promote of EM Cases. My funding organization ‘The Schwartz-Reismann Emergency Medicine Institute’ (SREMI), a non-profit academic institute who’s mission is to improve research and education in EM in Canada, also helps promote EM Cases through helping me with research projects related to EM Cases and through the conferences that they run. However, the most important aspect of promoting my podcast is being active and engaged in the EM community and in particular in the FOAMed world. Word of mouth is the most powerful promotional tool in my opinion.