Audio in 2015

Assumptions

• Mobile is now a first class object. People spend significant amounts of time on mobile and are capable of doing complicated tasks on their phones. Example: VSCO Cam, time spent browsing vine
• Due to early evolution of the podcast form factor, current technology is unable to evolve. The best that can be done is things like relevance algorithms/minimal network effect optimizations on the consumption side. Audio publishing technology is still in the era of RSS feeds when everything else has moved on to centralized distribution networks like Twitter.
• There are two major downfalls of audio as a form factor. The first is how information sparse it is compare to photo & video. This is exasperated by it being very difficult to go back and edit audio, a thing that is incredibly important for essays which is a media type similarly sparse to audio.
• The second downfall is the difficulty of consumption. This is also related to the first problem. Because audio test to be long & unedited (specially when done by amateurs) it requires a significant amount of investment from the person consuming it without any clear payoff.

The Solution

In order to evolve audio publishing past the current distributed protocol technology state it is in a completely new audio production client needs to be created. Thanks to how powerful mobile has become this client can now be a reality.

The production client needs to allow the user to record their thoughts without preparation or anxiety and then easily edit them down into publishable highlights. I believe a (very forgiving) ratio of raw recorded audio to interesting content is 12:1 (for every 12 minutes of recorded audio there is about 1 minute of interesting and consumable audio) this turns into 5 minute episodes for an hour of unedited material.

This editing will be closer to a selection of which clips to keep and look nothing like the current cut & paste audio editing we know today. The clips will be automatically broken up based on pauses. The resulting product will be harsher edits but much quicker editing: less than 5 minutes time to edit an hour of audio.

Also allowing for the creator to place “bumpers” or short music tracks in between sections helps elevate the final product and make it much more consumable. Think like what Mediums design and attention to typography does for the final essay.

The production client can bootstrap off of the current podcast infrastructure for publishing, allowing it to be built and released as a standalone client. But eventually the goal is to then create a dedicated consumption client. This is where the possibilities really become exciting. You can allow listeners to skip through sections & deep link to sections, allow paid subscription to people, detailed analytics, easy packaging for sharing “OneShot” like images of a specific section of audio, through channels & networks promote new content to people, and many more.

Prior Art

Multi Track DAW. A fully functioning multi track audio recorder and editor on the phone.

A podcast created by Slack which contains 7 short sections over 30 minutes. Included in one of these sections is a conversation with BJ Novak that is recorded on a phone on the sideline of a hockey game.

VSCO Cam

An incredibly thorough and detailed photo editing app which is very popular especially with teens.

This is a podcast I made with a friend of mine that follows the format described here. We recorded an hour of raw audio about a number of topics and then edited it down to 5 minutes and put in bumpers. That editing process took us about 3 hours (which in my research actually seems like a pretty short time)