Lessons from This American Life’s audio hackathon

Hacking together the future of audio

Screen Shot 2015-10-01 at 12.24.04 PM

September 19-20 marked the first ever hackathon to advance technology for audio storytelling. The two-day Audio Hackathon, sponsored by This American Life and hosted by Thoughtworks in New York, attracted diverse design, developer, and producer talent from around the world.

The Pop Up team was there providing support for our Audiosear.ch podcast API. Over the course of 48 hours, 12 teams created 13 proof-of-concept projects exploring what’s possible in audio technology. The questions that arose during the hackathon reflect many of the questions being asked of the audio industry today.

How can we make audio sharing more fun?

Screen Shot 2015-09-30 at 1.23.29 PM

Would listeners be more compelled to share audio if they had more creative control over their message? Projects like Earmoji and For Your Ears Only let users inject their own commentary into audio while sharing.

Earmoji created a set of audio responses with visual icons. Armed with a new vocabulary of “earmojis” from Snoop Dogg to the opening bars of the Serial theme, listeners can respond to audio in ways that are funny and infinitely shareable.
For Your Ears Only made sharing personal: users could share podcast episodes with a personalized audio message delivered via email to a person of their choice.

How can we better share audio with social networks?

Though much podcast audio may still be niche, it’s doesn’t have to exist in a closed ecosystem. Projects like Soundszr and Hearsay understood that you can attract new audiences by catching potential listeners where they’re already hanging out on social media. Soundszr helps cut and share audio segments on Facebook, and Hearsay lets you highlight text + audio moments (sound familiar?) on Twitter and Facebook.

In contrast, projects like Listening Party and Pod With Friends set out to create a new, social environment for listening.

How can audio play nice with other media?

Screen Shot 2015-09-29 at 11.44.13 AM

The Hearsay project recognized that audio on the web can be multimedia experience, testing a hypothesis that audio can become more engaging when combined with text and video. The team pulled transcripts from Audiosear.ch and used the text to make short GIF-like videos for audio segments, one word at a time.

The Soundscape project also looked at ways of integrating audio with other media by “synchroniz[ing] social and multimedia content with your audio stream. [Soundscape] incorporate[s] timed photos, tweets, videos, maps, and links to enhance the listening experience.”

How can podcasts reach new listeners?

Screen Shot 2015-09-30 at 1.21.19 PM

One recurring question over the weekend was how to make on-demand audio more mainstream. As one audio hacker put it: how can we “Beyonce-fy” audio?

When it comes to attracting new listeners to the podcasting medium, one group arrived at education as an answer. By signing up for the My Radio Class (at myraclass.com — say that one out loud) newsletter series, new listeners get a gentle onboarding to podcasting, straight to their inbox.

How can we identify shareable segments?

Screen Shot 2015-09-30 at 2.53.56 PM

Several projects at the hackathon, like Soundszr and SoundBits, looked into ways of creating smaller units of audio to share. In this way, listeners can react to audio in more specific ways by sharing of particular moments of interest rather than entire episodes or shows.

One question that arose was: who are the best people to make these segments? Should curation only be entrusted to the audio creator, or should it be easily accessible even to your average Joe/sephine podcast listener?

How can we make recommendations personal?

Screen Shot 2015-09-29 at 11.46.26 AM

For music, there’s Spotify and Pandora. For movies, there’s Netflix. But podcast recommendation is still fragmented: short of wading through an archipelago of podcast recommendation newsletters, there’s no reliable way to find audio for your mood and interests.

Projects like QueueCast and SoundBits took inspiration from the dating app Tindr, where users swipe right for a match, and left to pass. This user data, combined with data from other tastemakers, can help listeners discover compelling new audio just for them.

See all projects from the Audio Hackathon*

*Big thanks to Hearsay, QueueCast, Clipper, Boardcast, and Soundbits for their creative uses of the Audiosear.ch API!