Just over a week ago I wrote about a video experiment I’m involved in.  The experiment amounted to publishing a video version of an audio program published on the Just One More Book!! podcast website to see how an un-promoted video fared against its audio counterpart — a mainstay on the site.

Basically, I took about forty minutes to throw some still images into a thirteen-minute audio program and created a Quicktime movie which I uploaded to Viddler.  I then embedded the flash player for the video into the post.

The video was watched about three times more often than the audio was downloaded on the first day of the post.  Now, ten days later, the video has been viewed 2,100 times, slightly more than four times as often as the audio played/downloaded.  Aside from a mention in the post, we didn’t use any form of promotional campaign to raise awareness of the video.

We’ve received some feedback that there should be more visuals and more regular transitions in the video to keep the audience’s attention.  I completely understand that, particularly since I rushed the video together merely as a test.  The results are significant, though.  They prove that video players are far more obvious than their audio counterparts and that the general public understands video first and audio second — likely due to the success of services like YouTube.  As a result, we’ve moved the player for our audio programs (Podpress) to the top of each post and we have plans to do more with video.

I don’t believe in gratuitous use of video.  For example, exclusive use of video for a talking head seems to me to be a waste of bandwidth and offers little value to the viewer.  My intent is to use video to offer visuals that support and enhance the experience for the audience.  It’s more work for the content creator though it makes more sense to use audio for things to be heard and video for things to be seen.

There will be a discussion about this on next week’s episode (#121) of the Canadian Podcast Buffet.

PS. did you click on the image to see if a video or audio program would play???

Here is the long-overdue summary of things to consider when producing a story — in audio or video.

1. Be where the action is (when you audio/video record)
2. Capture the energy through ambient sound/video recordings
3. Set goals and be prepared to adapt them as you go
4. Gather and label your material
5. Understand that editing is an iterative process
6. Look for candidate themes, characters, plots and stories
7. Let the story identify and tell itself (if you decide the specific story in advance, you may find you’re wrong)
8. Tell the story using different voices and other elements
9. Remember that you can tell as much of the story with what you cut out as you can with what you include
10. Avoid soundbites
11. Experiment with recording, editing and storytelling
12. Only use your own voice as much as it’s needed
13. Use music and ambient sound/video as a backdrop and for transitions
14. Listen/watch, edit, mix, repeat
15. Consult with a set of honest (even blunt) ears/eyes that aren’t attached to your own body
16. Keep your listener/viewer wanting more at each moment
17. Don’t crowd your listener/viewer
18. Be patient
19. Enjoy (even be proud of) the process and the results

I realized that the missing element was music.  When my feelers for original music came up empty, I went to the Jazz collection and found the perfect piece of music: Poor James from the CD Tractor Parts: Further Adventures in Strang by the Canadian band, Zubot and Dawson.  Sadly, the band is no more.  It was Ken Rockburn who introduced me to the band when he played their fantastic song, Tractor Parts, on All in a Day back in 1998.

Once I picked the music and started placing it, things started feeling right with the show.  I had wondered if I should include a second piece of music.  That can be too much, though, and I decided to limit myself to the one song.  I also decided that it would be used specifically to score the excerpts of the reading.  There has to be logic to the use of the music, it can’t be random.

The editing and refining continued, though most of the effort was in cleaning up edits, removing some bumping of the microphone and, in two cases, cleaning up mistakes made by the speakers.  Thankfully, the way each speaker corrected themselves facilitated a clean edit and you’d never know.  Purists would argue to leave the mistakes in.  However, in this particular show, it made more sense to clean up the mistakes.  They errors were minor enough and the responses to the mistakes more intrusive than the mistakes themselves.

OUTSIDE EARS

I did a thorough refinement of the first four minutes of the show and listened to it for both audio and story flow.  It still didn’t feel right.  When I played it for Andrea, she pointed out a few things that didn’t work for her.  In particular, my voice over, the readings and interviews had different energy and emotional levels.  The readings and the interviewee spoke in more bursty and extreme intonations and my voice over was mellow and reflective.

MOOD AND ENERGY FLOW

I re-recorded my narration to be more punchy and energetic and played back a rough mix.  That was much better.  The individual clips made more sense as a unit so I went back to smooth out the hand-offs between clips and mix the music.  I also boosted the level of the ambient sound during my introduction so that there was more energy in the opening.

REFINING THE MIX

The process of mixing the elements for a smooth flow seems to me like staining a quality piece of furniture; it takes several passes, patience and a lot of care.  I worked on that for a while and realized that many of the voices on the show seemed a bit sharp in the mix.  I’m not sure if that’s the mic or the qualities of the voices.  I used my equalizer to add some warmth and remove some of the clarity from recordings.  I should note that I hadn’t adjusted any of the frequencies prior to that.  EQ’ing is the last thing I do.  I only do it to make sure that each element has its own room to breath.  If you find that two pieces of mixed audio are fighting with each other spend less time with the volume faders and more time giving each piece of audio its own space in the audio spectrum.

FINAL PRODUCT

I finished the audio production work in the early evening and created a mix I was happy with.  All it took was patience, listening, feel and an iterative approach to finding the pieces that best told the story.  In the end, I went from seventy-eight minutes of recorded audio of the event to a nine-minutes and fifty-four seconds of reading excerpts and interview clips included in the show.  My entire voice-over of introduction and narration was one-minute and seventeen seconds.  The program is thirteen-minutes and sixteen seconds long and has been published under the name A Bear in War on the Just One More Book website.

 

 A Bear in War [13:16m]: Play Now | Play in Popup | Download

SOMETHING DIFFERENT

I’m an audio guy and I often profess that audio podcasts are the best option for mass distribution and ease of consumption.  However, even I can’t ignore the popularity of video.  More to the point, no matter how hard we audio folk try to make it obvious that site visitors can press play on an audio player, many people just don’t get it.  Even if it is far more intrusive (that is, you are glued to your video screen to consume a video), video on the web is far more obvious to use.

So, we’ve tried an experiment.  We published the “A Bear in War” documentary as a normal audio podcast of Just One More Book and to help people understand that there’s multimedia content available on the site, I created a video/slideshow version of the documentary (using the same audio), uploaded it to Viddler and embedded it within the post for A Bear in War.  Given more time and energy, I would have done more to make the video more compelling for the average viewer by using transitions, pans and zooms — like Ken Burns.  I threw the video together very quickly.

Interest in the video is encouraging.  After about two hours, the unpromoted video was watched thirty-six times.

FINAL TALLY

Not including travel time and the blogging I’ve done to document my workflow, but including the research and recording times and the production of the video, it took me about sixteen hours to create the A Bear in War documentary.  It’s not unusual for a documentary to take a significant amount of time to produce which is why I don’t do many of them.  I do very little, if any, editing in most of my hobby production work which means that the average ten minute podcast probably takes about thirty minutes to record, produce and upload.

I’ll create a summary of my workflow post highlights later today.

A couple of quick points I thought I’d add to this…

The “A Bear in War” documentary program will be published at 1:00am, Monday, November 10, 2008.  That means the total time I have to work on the project from once I hit record is about 32 hours.  So far, my time investment so far has been as follows:

• Advanced Research: 90 minutes
• Recording: 180 minutes (accounts for the entire time I was at the museum)
• Step One (gather and label material): 10 minutes
• Step Two (listen to and edit audio, iteration one): 100 minutes
• Step Three (look for candidate themes and plots): 30 minutes
• Step Four (select keepers…, iteration two): 15 minutes
• Writing first draft of the opening narrative while lying in bed: 15 minutes
• Blogging my workflow: 40 minutes
• TOTAL INVESTMENT SO FAR: 8 hours

All of the recording equipment I needed for the event, along with some extra cables and batteries, a camera and a paper-based notebook fit in a single hardshell case which I was able to carry with me as I walked to and from the War Museum (roughly 40 minutes each way, not counted in my time above) to record the event.

I’ve noticed that workflow and media production methodology is rarely talked about at Podcast and new media conferences.  Also, a few people have asked me how I approach my own audio/video recording, editing and mixing to tell stories — sometimes with a narrative, sometimes without.

I’m actually working on a project right now.  So, the timing is right for me to document what I do.  Depending on feedback, I may consider this as a presentation at a future conference.

BACKGROUND

I attended and audio-recorded the launch of the children’s picture book A Bear in War at the Canadian War Museum this morning.  The book tells the incredible and true story of a stuffed bear that was mailed to a Canadian soldier by his daughter during World War I.  The soldier, Lt. Lawrence Browning Rogers, volunteered when he was thirty-six years-old (a “greybeard”) and was sent to Europe where he served as a medic in Belgium.  Lt. Rogers and his family corresponded regularly.  To help her father feel connected with home, Aileen sent her beloved “Teddy” to the front lines where it kept Lt. Rogers company for about a year-and-a-half, until he was killed in action on October 30, 1917 in the Battle of Passchendaele.  Teddy was shipped back home with Lt. Rogers’ other personal effects and military honours.  In 2002, Teddy was donated to the Canadian War Museum and is now an attraction of the World War I exhibit in the Gallery.

MATERIAL

I gathered about seventy-eight minutes of audio including the welcome speech by a museum curator, the book reading by the two authors, a brief Q&A session, some ambient sound of the event and six interviews.

THE GOAL

Tell a compelling story in roughly fifteen minutes.

STEP ONE: GATHER AND LABEL MATERIAL

I like to do things in an organized and logical manner.  The first thing I do is create a new project in my production software.  I use Cubase for all of my audio production work.  Once the project is created in its own folder on my hard drive, I import all of the source audio I have collected and I make sure that each imported file has a descriptive file name.

STEP TWO: LISTEN TO AUDIO AND EDIT (ITERATION ONE)

A lot of people map out, even storyboard, their media productions before they do anything with it.  I don’t.  In fact, to me, the story that I want to tell isn’t always obvious to me.  That is, even though I may have an idea of what I want the result to be, I find that my ideas evolve or even radically change once I start listening to the individual pieces and discover that they can be threaded together in an entirely different way to tell the same story.  Sometimes, the original story idea turns out to be terrible compared to what can be done.  So, I approach this in an open minded way and eliminate what I determine I don’t need rather than look for the stuff I want to keep.  Having said that, there are usually some clips that are quite obviously gems.

Okay… even the gems sometimes turn out to be duds.  But if you discover what you think is a gem, label it that way so you know.

All that to say I approach editing as an iterative process and, at this stage I do a simple straight-through listen.  I don’t dig in too deep or replay clips.

As I listen during the first pass, I make cuts in the audio to divide and identify specific pieces of audio.  In an interview, that would be a question.  In ambient noise, it may be a particular sound or event that has resulted in some cool audio.  For this audio program, I want to use some of the book reading so I’ve made some cuts where a particular compelling moment of the book is being read.

I use colour coding to help me identify and group related elements.  In some projects I colour code based on speaker.  In others, I’ll colour code based on theme.  In the “A Bear in War” project I’m colour coding based on known keepers.

STEP THREE: LOOK FOR THE CANDIDATE THEMES, PLOTS AND STORIES

I try not to limit myself to a single story line in my media work.  Life isn’t made up of single themes, plot elements and stories.  It’s a spaghetti of stories and ideas that weave around each other.  As long as I can travel with my listener/viewer along an arc,  I like to let a few of these stories play together.

In this step, I look at the different keepers and labels to see if anything jumps out and me and, more importantly, that there is enough connected audio that can tell the story in an engaging way.  Engaging means knowing what to include and what to keep out.  If you overdeliver the story, your audience won’t have a chance to use their own imagination.  If you underdeliver, they won’t have enough material to engage their imagination.  It’s a balance and I’m not sure that I’ve figured out how to hit the sweet spot, yet.

For “A Bear in War”, I know that Teddy is going to be one of my main characters.  That’s an obvious one which can be dangerous, too.  However, since there’s surprisingly little coverage of this in the mainstream media, particularly in the children’s book world, it’s a freebie.  It’s also universal.  Who couldn’t relate to the importance and value of a cuddly toy to a child and their parent?  Themes I’ve identified include the use of Teddy as the narrator in the book, crafting the voice of a teddy bear, connecting world conflict in 1917 with world conflict in 2008, the human factor in war time and engaging with children on an important topic (rather than pretending it doesn’t exist).

STEP FOUR: SELECT THE KEEPERS AND REMOVE ELEMENTS THAT DON’T SUPPORT THEM (ITERATION TWO)

It’s time to make some tough decisions.  The important thing to remember is that your audience doesn’t know what you cut out, only what you deliver.  So, pick the stuff you can tell well and get rid of any content that doesn’t help you deliver.

Sometimes you have to sleep on these decisions.

I think that’s what I’ll do.

Based on a number of conversations I’ve had recently, I’ve decided to host a webcast on audio mixing.

This webcast will explore:

• audio mixing for creative effect
• how audio frequencies can impact your mix
• using your ears to mix for your listener

Instead of demonstrating and supporting specific software packages, I will be discussing concepts and providing overall guidance.

The event is free and limited to 15 people. Because space is limited, be sure to attend if you do register.

Click here to register.

Please let me know if there are any other topics you’d like to be considered for future webcasts.

Part 4 of this series dealt with the creativity of editing interviews. In this installment, I am returning to applied editing.

When I first mapped out this series, I figured I’d be done by now. I had a neatly compiled list of the types of edits I do and some soft skills I wanted to share. As it turns out, I left out a number of really interesting ideas and techniques - so the series continues.

One of the techniques I had overlooked is one I’ve dubbed Buried Edits. It’s an easy one to forget because I use it so rarely. However, it is a very powerful way to unite two different recorded elements, and disguise the edit in the process. For the purposes of this entry, I will present in terms of spoken word recordings even though the same concepts can apply to music - or any other recording.

First of all, what’s the trick?

In a Buried Edit, I unite the same sound that appears in two different recorded takes using a Clean Edit.

What???

The spoken delivery of two recorded elements may have been significantly different in each take while still having a common delivery element in each. I call this common element the “pivot”. In a Buried Edit, you find a pivot inside a particular word in each take and make your cut inside these pivots, then marry the two desired pieces together.

An example, please.

The word “so” is a really good word for a pivot. The “ess” sound is nice and static. You can cut that sound in half from two different sentences and marry them in a way that noone would ever know, as long as the audio quality is similar, and the thought is continuous. Even if the delivery by the speaker has a different cadence in each sentence, it’s a very forgiving sound.

When to use Buried Edits

In all cases, the speaker in the recording I am editing has run words together which makes it difficult to have a transparent cut between words. I have used Buried edits in the following situations with great success:

• The speaker has gone on a tangent and I need to remove a large section of talking. I find a pivot within each of the first and last thoughts and eliminate the thought process. Having a pivot means that I was able to make the speaker’s comments more efficient and fluid.
• The speaker records several takes in a row, sometimes without a pause. Different elements of each take are great while each full take has its problems. I will take the elements from each take that work well and make them into one stream. Buried Edits work well here because it isn’t the insides of the words that pose a problem, it’s the first and last sounds of the words.

In part 6 I’ll talk about something I call a Staggered Edit to disguise an edit in a multi-track recording of two speakers.