NEH Awards $235,000 to Improve Accessibility to Historical Audio Files

A professor in the School of Information at The University of Texas at Austin has been awarded a National Endowment for the Humanities (NEH) grant to develop the tools to sort and analyze recordings of poetry, speeches and folk tales in much the way that music files are already routinely analyzed.

The work could make hundreds of thousands of audio files more accessible to scholars and researchers.

The research by Assistant Professor Tanya Clement is, in part, a response to an August 2010 report by the Library of Congress and the Council on Library and Information Resources titled, "The State of Recorded Sound Preservation in the United States: A National Legacy at Risk in the Digital Age" that says our culture will not preserve what it does not know how to use.

According to Clement, libraries and museums across the United States have tremendous collections of audio files that are, for the most part, inaccessible.

"Computer scientists have created tools that group music according to such things as pitch, tone and tempo in order to categorize them as jazz, pop or classical," Clement said.  "Similar systems have not been developed to do the same kind of analysis on spoken text files, one factor that makes poetry, folklore, or presidential speech recordings unsearchable and therefore, inaccessible."

The funds will be used to support a four-day institute in Spring 2013 at The University of Texas at Austin geared to humanities scholars, librarians, curators, collectors, computer scientists and archivists.

The four-day workshop, called the Institute for High Performance Sound Technologies for Analysis and Scholarships (HiPSTAS), will focus on developing tools to help analyze speech, using criteria such as patterns, pitch, spectral range and speed of speech. The attendees will focus specifically on poetry, speeches and folktales.

One year later, a follow-up workshop will focus on implementation of the analytical tools and advanced technology (classification, clustering and visualization) to develop a suite of digital tools for use by interested institutions.

"There are hundreds of thousands of important spoken text audio files, dating back to the nineteenth century and up to present day that represent significant literary figures and bygone oral traditions that are virtually inaccessible in this digital age," said Clement. "We have very little access to these cultural artifacts."