News
November 9, 2010

WGBH Works with Nuance Communications, Inc. on Effort to Improve the Quality of Live News Captioning

Boston, Mass., – November 9, 2010 – WGBH's National Center for Accessible Media (NCAM) is collaborating on a U.S. Department of Education-funded Project to develop a prototype system to automatically assess accuracy of real-time captions for live news programming. The system will identify and weigh the impact of different kinds of errors in closed captions.

The variations in quality and accuracy of real-time captioning have long been a concern to people who are deaf or hard of hearing. Captioned news in particular can be a lifeline to viewers who cannot hear well enough to follow program audio, a lifeline only as strong as the captioning is accurate. Live, or real-time captions are most often produced by specially trained court reporters, called stenocaptioners, who use a phonetic keyboard attached to a computerized dictionary to translate spoken words into text in a matter of a few seconds. National consumer advocacy organizations have petitioned the FCC to consider adding a standard of caption quality to existing mandates for quantity of captioning on television.

An early activity of NCAM's "Caption Accuracy Metrics" Project was the design and launch of a web-based national consumer survey of television news caption viewers. Respondents were asked to read sample captions and identify error types that make captions hard to follow. Basic real-time captioning error types are word substitutions, deletions (dropped words or omissions) and insertions (addition of words not spoken).

WGBH is working with Nuance Communications, a leader in speech solutions, to develop customized language processing, data analysis, and benchmarking tools for the Project. Nuance is leveraging its Dragon NaturallySpeaking speech recognition technology to develop these tools in order to quantify and further analyze the captioning data that is being gathered. Nuance's work is integral to the Project's development of a final prototype tool that will enable tracking of audio and captioning, comparing the spoken word with the caption output and rating the accuracy levels based on error type and severity.

In addition to software developed at WGBH, The Project is using Nuance’s Dragon speech recognition as a benchmarking and data analysis tool. The Project is not using speech recognition as a caption-creation source. Test data reports will not associate samples with specific television networks, programs or caption agencies.

Project staff are working with a team of advisors, including the National Institute of Standards and Technology (NIST), the federal technology agency that works with industry to develop and apply technology, measurements, and standards, Gallaudet University and the National Technical Institute for the Deaf.

NCAM staff recently convened a technical review panel of major stakeholders in caption quality at the CBS Broadcast Center in New York. Representatives from broadcast and cable networks, caption agencies, deaf education experts, and the National Court Reporters Association discussed the value of the project’s work to their organizations. In addition, the FCC recently announced that it is refreshing the record of comments on caption quality, solicited in two notices of proposed rulemaking over the past five years, and will determine with consumer and industry feedback if standards of quality should be set and, if so, in what manner.

"The phrase 'a rising tide lifts all boats' works well here," said NCAM project director Marcia Brooks. "Being able to quantify caption errors has been the missing piece for a range of stakeholders. We expect the Project will help improve captioning quality through better and easier monitoring by caption agencies and networks."

Funding for NCAM's Caption Accuracy Metrics project is provided by a grant from the National Institute on Disability and Rehabilitation Research, U.S. Department of Education under grant #H133G080093-10.

About NCAM and WGBH

The Carl and Ruth Shapiro Family National Center for Accessible Media at WGBH is a research, development and advocacy entity that works to make existing and emerging technologies accessible to all audiences. NCAM (ncam.wgbh.org) is part of the Media Access Group at WGBH, which also includes The Caption Center (est. 1972), and Descriptive Video Service® (est. 1990). For more information, visit The Media Access Group at WGBH, access.wgbh.org.

About Nuance Communications, Inc.

Nuance (NASDAQ: NUAN) is a leading provider of speech, imaging and customer interaction solutions for businesses and consumers around the world. Its technologies, applications and services make the user experience more compelling by transforming the way people interact with information and how they create, share and use documents. Every day, millions of users and thousands of businesses experience Nuance’s proven applications and professional services. For more information, please visit: www.nuance.com.

Press Contacts:

WGBH Contact: Mary Watkins
mary_watkins@wgbh.org, 617 300-3700

Nuance Contact: Lisa Mokaba (InkHouse)
lisa@inkhouse.net, 781 791-4570

###