Academic journal article American Annals of the Deaf

Evaluation of Captioning Features to Inform Development of Digital Television Captioning Capabilities

Academic journal article American Annals of the Deaf

Evaluation of Captioning Features to Inform Development of Digital Television Captioning Capabilities

Article excerpt

With rapid advances in digital technology, the broadcast television industry will soon be providing cinema-quality sound and video in the home. Given this technological advance, how should captions change to better serve consumers who are deaf or hard of hearing? To address this question, consumer preference data were collected from 207 deaf students and adults. Videodisc examples of various captioning feature combinations were used to obtain ratings from these consumers. Findings from multivariate conjoint analysis show that some current captioning features should be retained (white letters and speaker-dependent placement), while other features should be adopted (captioning in sentences and use of a clear or semitransparent background). These findings offer guidance to the captioning and television industries as they launch the digital television age.

Prior research on captioning has investigated various display features and their benefits for students who are deaf or hard of hearing (Appell & Byrom, 1993; King & LaSasso, 1994). Findings have supported the use of mixed uppercase and lowercase (i.e., mixed case) letters; increased use of speaker identification and sound effect cues; and the use of color to highlight nondialogue audio. However, these studies generally have limited their choices of features (treatments) to those supported by existing analog television decoders, whether built-in or set-top.

With the advent of digital television, however, many of the constraints on captioning's display features will be removed. For example, the bandwidth for character transmission will be increased from the current 480 bits per second (bps) to 9600 bps (EIA 708, 1997). This will enable full-verbatim captioning, when feasible.1 In addition, the basic caption character font could be changed from the current dot matrix characters to fully formed characters that would be more readable and attractive. As a consequence of the Television Decoder Circuitry Act of 1990, all digital televisions must have built-in decoders.

Notwithstanding the myriad technical possibilities offered by digital television, many questions regarding captioning features remain unanswered. Should captions present whole sentences, or should they continue to present sentence "fragments" as is the current industry practice? (The term fragments is used because the parsing of the dialogue into segments that are displayed as discrete captions is not always based on thematic, semantic, or syntactic units.) Should captions continue to be composed of white letters in a black box, or should they be changed to a style and conformation approaching the appearance of subtitles? Should captions be white or in color? Should captions be placed on a transparent, semitransparent, or opaque background? The fundamental question is: What do captioning viewers want?

Captioning Features

Captioning features can be divided into two broad categories: stylistic and technical.

Stylistic Features

The current style of captions is affected by the limitations of analog television technology and decoder capabilities. Within these technical limits, various captioning practices and policies evolved, creating a "style" of captioning that utilizes a subset of the features provided by the technology. As Table 1 illustrates, there are several options that relate to these matters of style that could be varied to best meet the needs of people who are deaf, hard of hearing, learning English, learning to read, or otherwise able to benefit from captions.

Current standard captions convey sentence fragments with near-verbatim content and speaker-dependent (variable-position) horizontal and vertical placement (although placement at the bottom of the screen is typical). Each caption is limited to four lines, but the actual number of lines varies depending on the number of words and the use of placement to convey speaker identification. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.