International Captioning Project
Pre-recorded captioning
Off-line captions are created for non-live, pre-produced programming such as dramas, documentaries, or instructional videos. These captions are also known as "pop-on" captions, because of the way they appear on the screen, in the manner of foreign-language subtitles. Unlike subtitles created for foreign-language movies, though, pop-on captions are also be used to display non-speech information, such as sound effects or speaker identification. Pop-on captions may be placed virtually anywhere in the video picture to indicate who is speaking, or from where important sounds are emanating.
Using a cassette work copy of the original (uncaptioned) master video tape, a caption writer (also known as the caption editor or subtitler) transcribes the audio portion of the show into a PC-based word processor running special captioning software. The caption writer uses the software to assign frame-accurate timecodes to each caption, dictating when the text will appear and/or disappear from the screen, and also assigns specific placement codes to each caption, indicating where the text will appear on the viewer's screen. Depending on the program being captioned, the off-line captioning process can take up to 15 hours per broadcast half-hour.
One additional function of the caption writer may be to edit the text for reading speed and, depending on the type of program or even on the captioning agency itself, for language, as well. There are, however no regulations governing editing style. While verbatim captioning is generally professed to be the goal of the captioning service, it is usually agreed upon that verbatim captioning can, at times, be too fast for many people to read and understand. However, there is often disagreement on how, or when, to edit captions. Some caption providers limit themselves to removing fillers, such as "uh" and "ah", but even this can cause problems if such editing actually alters the "flavor" of the character speaking the words. Many service providers will edit to a certain reading speed while not simplifying the language. (Exceptions occur when captioning children's programming; in this case, captions are often edited for both reading speed and language.) Some agencies edit captions for all programs, regardless of the intended audience.
While all captioning agencies produce captions which look relatively the same, the lack of regulations governing caption style means that the end result will vary slightly from agency to agency. For example, one caption provider may choose to delineate changes in speaker by positioning caption text directly below the character speaking; another may choose to place all captions at the bottom center of the screen, delineating speaker changes by using a pair of dashes before each caption. Still another may use a different color for for each new character. (Teletext countries have had access to color captions since the system's inception, and the use of color to indicate who is speaking is common. The United States, on the other hand, has had the capability to use color but until recently, decoders were not available to decode color captions. It has yet to be determined how to make best use of color capability. While all teletext decoders can display color, it is an optional feature in built-in line-21 decoders and therefore must be used judiciously.)
After a final review, the caption text file is converted into coded data and sent to a tape or broadcast facility for encoding. In North America, the encoding process usually occurs prior to broadcast. The caption data are loaded onto a PC at the tape facility; the master videotape is played back through an encoding device, with the timecode track from the master videotape being fed to the PC. The timecode triggers the caption data, which are sent to the encoder and inserted into the video on line 21 of the vertical blanking interval. The newly captioned video emerges from the encoder and is recorded onto a second videotape, which is called the closed-captioned master. Once the encoding session is finished, the closed- captioned master is ready for re-broadcast, duplication or distribution, the language stream (or streams) are now permanently embedded on the tape.
In Europe and Australia, subtitles are actually transmitted at the time of broadcast, rather than encoding them prior to air. Using technology similar to that used in pre-broadcast encoding, the data are copied onto a floppy disk, which is inserted into a PC at the broadcast site. The timecode track from the master is fed to the computer, triggering data to be sent to the encoding device. Data and video are merged in the encoder and sent to the transmitter. Because the subtitle data are not recorded onto videotape, each time the program is broadcast the subtitles must be retransmitted.
