You are here

Helping Google (and scholars, researchers, educators, & the public) find archival audio

  • Anne Wootton,, Pop Up Archive

Culturally significant digital audio collections are hard to discover on
the web. There are major barriers keeping this valuable media from
scholars, researchers, and the general public:

Audio is opaque: you can’t picture sound, or skim the words in a
recording. Audio is hard to share: there’s no text to interact with.
Audio is not text: but since text is the medium of the web, there’s no
path for audiences to find content-rich audio. Audio metadata is
inconsistent and incomplete.

At Pop Up Archive, we're helping solve this problem making the spoken
word searchable. We began as a UC-Berkeley School of Information
Master's thesis to provide better access to recorded sound for audio
producers, journalists, and historians. Today, Pop Up Archive processes
thousands of hours of sound from all over the web to create automatic,
timestamped transcripts and keywords, working with media companies and
institutions like NPR, KQED, HuffPost Live, Princeton, and Stanford.
We're building collections of sound from journalists, media
organizations, and oral history archives from around the world. Pop Up
Archive is supported by the John S. and James L. Knight Foundation, the
National Endowment for the Humanities, and 500 Startups.