novel methods for splitting, organizing, & transcribing interviews (podcasting)

it may not be the first choice people think of for podcast production software, but there is no daw as flexible, fast, or intuitive as ableton live for any kind of audio work. when creating a podcast, interviews are usually at the core of creating a strong piece & it's essential you have some means of keeping these organized so you can find & pull out tape as it's required in your production. typical professional practice involves transcribing & time coding the whole thing but this can be extremely time consuming when you're working fast on your own without staff. here instead I present my idiot-proof method for organizing interviews in a fast & highly useable way. 

chopping up the interview

typically, interviews will range between 45 min to an hour & a half which is totally unmanageable to seek through. coming from a music production side, I tend to prefer to work from smaller samples & the same philosophy can be applied to podcasts. quick note before we begin - make sure "warp" is turned off in your clip, especially if you imported the audio.

start by just listening through & cutting every time the subject changes - this usually happens between questions but occasionally the interviewee will diverge topics in their responses. click on the clip's waveform to place the insert marker where you want to cut & use the key command cmd + e to split the it there.

"warp" decouples your audio's pitch & speed. this could result in distortions in your interview, so make sure it gets turned off

"warp" decouples your audio's pitch & speed. this could result in distortions in your interview, so make sure it gets turned off

splitting & renaming clips in ableton live for podcasts

after that, select the new clip by clicking on the color bar above the waveform & press cmd + r to rename it. start with a sequential number (1. , 2. , 3. , etc.) so you can keep everything in proper order later. when you name it you'll want to make sure it's brief but descriptive of the contents for quick reference.

time-wise, a general rule of thumb to go by is to not cut a clip shorter than 30 seconds or longer than about 2 minutes. there are exceptions to this, as with anything, but the point is to cut the interview down to digestible components. 

clip collection

once everything is split in the project you need to collect them into files. it's in no way a daunting process - simply highlight all the clips, right click, & select "crop clip(s)." this will commit the size & names you've assigned to each clip.

crop clips ableton live podcasting

now look over at the browser & find "current project" in the sidebar. navigate through the folders: samples -> processed -> crop. et voila, here's your interview, chopped & labeled.

browser ableton live podcasting

finally, right click the crop folder & select "show in finder" from the context menu. from here, you can drag the folder to wherever it will live - a flash drive, documents folder, wherever - & rename it to something more useful than "crop."

loose transcription

we have to listen more or less in real time, but reading & scanning can be non-linear which works much faster. therefore, putting text to your audio is massively helpful when seeking out tape as you put your story together. how you go about doing this exactly isn't a perfect science, but here's how I like to do it: 

open up the folder you put the clips in & you'll see there are two files for every subject. this is because ableton creates an .asd analysis file to help it keep track of what's in the audio file but you won't need these. the quickest method for getting them out of the way is to just "arrange by kind".

next, load up your favorite word processor & put it side-by-side with your clip list. type out all of the clip names & leave a bit of space between them - a paragraph or two will do.

"arrange by kind" will sort the .asd from the audio files.

"arrange by kind" will sort the .asd from the audio files.

now you'll want to open your clips in quicktime. if this isn't the default application for the file type, right click & select it from the "open with" sub menu. I like to just pull them up at once so I don't have to worry about doing a lot clicking & opening.

play each clip & type what you hear as fast as you can. don't worry about typing everything or spelling correctly, just grab key phrases & put down as much as you can. start a new line for every new idea. if you encounter an extended pause, denote this with "--" on its own line. 

transcription podcast ableton live

it's not beautiful, but this should give you enough to work with when you need to find ideas in your tape. what I've found is transcription is a classic case of diminishing returns - more & more effort yields less & less usefulness as you approach completion. it's the kind of thing you'll need to find your own comfort level with - if you can spend half the time to get 90% of the utility of a full transcription, is it worth it to you to double your effort to get the last 10%? as you work, you'll figure out what you need in your tape reference that's helpful to you.

better transcription - alternative

if you require a complete transcription or aren't a very fast typer, ableton once again offers the solution. select all of your clips & turn on warp. now you can lower the tempo (in the upper left corner). this will slow down the audio, giving you more time to type. another useful tool - select any clip & use the shortcut cmd + l to loop it, letting you listen to the selection over & over until you get everything down.