OLYMPIA - The Washington Office of Secretary of State’s Digital Archives, in a partnership with Microsoft Research, today announced it is hosting a new speech-search technology that will dramatically change how Washingtonians are able to access important public recordings.
The Washington Digital Archives is the first government program in the country to offer Microsoft Research’s Audio Search technology, which takes record keeping to the next level: it doesn’t just preserve audio recordings – it gives people an innovative way to search through them.
The partnership grew from an ambitious project undertaken three years ago by the Washington State House of Representatives and the Washington State Archives to save over 30,000 cassette tapes of committee hearings.
Secretary of State Sam Reed praised the Audio Search pilot project which offers the potential to increase the availability of additional public audio content.
“This is not just an enormous breakthrough for Washington State record keeping, it is the debut of a groundbreaking way to access public information,” said Reed. “The House cassette recordings were boxed up and losing quality in storage. Not only have these records been rescued, we are introducing a cutting-edge mechanism for the public to access them.”
Cassette tapes have approximately a 25-year shelf life before losing quality, and are the worst format for long-term preservation, according to Washington State Archivist Jerry Handfield. Typically, cassette tape content is preserved by transcription – someone listens to the entire tape and writes down the audio verbatim on paper.
Worried about the loss of information, the House of Representatives began a major project with the State Archives to convert their at-risk cassette tapes into electronic format in 2005. It made the files available to the Secretary of State’s Office and Microsoft Research, whose partnership is making it possible for the public to go to the Digital Archives Web site and search through words being said during the recorded hearings.
“By applying Microsoft Research speech search technologies to this innovative project, we’re allowing citizens to selectively access audio content that is important to them,” said Gail Thomas-Flynn, General Manager of State and Local Government for Microsoft. “Washington State is at the forefront of unlocking digital archives for all people to access.”
Visitors to the Digital Archives Web site can currently search through nearly 6,000 hours of House committee hearings from 1973 to 2001. For example, typing in “water” as a keyword in the search tab brings up all hearings in which water was talked about. This makes it easy for a user to speed through the hours of audio and go directly to a conversation point of choice.
Over the next two years, the House will add approximately 10,000 more hours of audio to the Digital Archives database.
“This process is going to be a breakthrough for legislatures everywhere to improve public access to invaluable records,” said State Archivist Handfield.
Since launching in October 2004 as the first of its kind in the country, the Washington’s Digital Archives has blazed a trail in the electronic preservation of documents. Washington is the lead partner in a multi-state project funded by the Library of Congress to demonstrate the requirements of building a digital archive system in states around the country.
More than 74 million items can be accessed online at the Digital Archives. Visit http://www.digitalarchives.wa.gov for more details.