CLOSED CAPTIONING FUNDAMENTALS
An In-Depth Guide
-presented by-
LINK Electronics, Inc.

THE NEED FOR CLOSED CAPTIONING:

Two important pieces of legislation, the Americans With Disabilities Act (ADA) and the Television Decoder Circuitry Act (TDCA) have made closed caption encoding and decoding a necessity for many video professionals. The basic mandate of the ADA is:   disabled people (including deaf and hard of hearing people) should have "an equal opportunity to participate in, and enjoy the benefits of, a service, program, or activity conducted by a public entity." Private companies are also covered by some provisions of the ADA. Accordingly, all educational and training videotapes and Public Service Announcements must be closed captioned, thereby providing deaf and hard of hearing people with the "equal opportunity to participate." (An estimated 5 million private companies will be affected.) The TDCA specifies that all television receivers "manufactured in the United States or imported for use in the United States" with screen size greater than or equal to 13 inches will have built-in closed caption decoders. This mandate was effetive July 1993. For the video professional, this means that the viewing audience of closed captions will vastly increase.

In addition, closed captioning benefits not only the estimated 24 million Americans with hearing disability but also the approximately 27 million in the U.S. who are learning English as a second language, another 27 million functionally illiterate, and 30 million elementary school students learning to read.

HISTORY OF CLOSED CAPTIONING:

Initial work on a closed captioning system began in the early 1970's with the Public Broadcast Service (PBS). Using a concept first proposed by the National Bureau of Standards, PBS developed and demonstrated the feasibility of a closed caption data transmission system. (With the cooperation of the National Association of Broadcasters and funded by the Office of Education, the Public Broadcasting Service was able to evaluate at least two systems for caption delivery.) By November 1975, PBS was able to submit to the Federal Communications Commission (FCC) a Petition for Rulemaking, asking the commission to amend its television standards to provide for the transmission of closed caption information.

This amendment was issued by the FCC in December, 1976, with an effective date of March, 1977. The way was now open for the full development of a closed caption system that went from feasible and desirable to practical.

The PBS "Captioning for the Deaf Project" began in June, 1976, with a number of sub-projects. Among others, PBS would "support the establishment of a captioning entity" (which turned out to be the National Captioning Institute), develop a commercial decoder/receiver for home use, and "provide technical assistance for the start-up of captioning activities by commercial networks".

The development of closed caption encoders began with a "Request for Proposals" issued by PBS in December, 1978. After evaluation of the responders, contracts were issued to Industrial Sciences, Inc.(?) and EEG Enterprises. PBS worked closely with the contractors and monitored the development phase of this program. (The EEG Simple Encoder and Smart Encoder were developed at this time.)

"After interviews, discussions and negotiations" with a number of companies in the consumer electronics industry, Sears, Roebuck and Company was targeted as the sales and service organization that would market the closed caption decoder for home use. The actual manufacturer of these decoders was Sanyo Manufacturing Company, already the major supplier of television products for Sears. In January, 1980, Sears' Marketing Public Relations office issued a statement entitled "New Telecaption Adapter Adds Words to TV Pictures". The Telecaption I decoder appeared in the Spring 1980 catalog for $250.00.

Also as part of this project, 20 "Caption Editing Consoles" were built by the PBS. Eighteen of them were delivered to the National Captioning Institute (NCI) in late 1979. By March, 1980, all consoles were in operation. This date coincides with the official inauguration of captioned program service by the PBS and the ABC and NBC networks.

MODERN CLOSED CAPTIONING, TECHNICAL ASPECTS:

Closed caption information is added to Line 21 of the Vertical Blanking Interval (VBI). It may be added to either or both the odd and even fields of the television signal.  However, the primary language program related information appears in the odd fields. (It should be noted that caption industry people typically refer to the odd fields as "Field 1" and the even fields as "Field 2".) To assure adequate caption performance wherever a usable picture can be obtained, a low instantaneous data rate of 503 kilobits/second (32H) was chosen. Data is preceded by a seven-cycle sine wave similar to color burst (called the "Clock Run-In") and three "start bits" that are always "0", "0", and "1".

Two bytes of data, using seven bits, odd parity format, are possible on a given line. The rise time is controlled (2T) and the amplitude of data and clock run-in is 50IRE units. Using only Line 21, Field 1, of the VBI, a delivery rate of about 3600 characters or 500 words per minute is theoretically possible (depending upon the particular caption style, discussed later).

Tests conducted by PBS determined that the typical reading rate for captioning is about 125 words per minute. (Incidentally, spoken dialogue may exceed 200 words per minute.)

Particularly with the availability of Field 2, the data delivery capacity (or "data bandwidth") far exceeds the requirements of simple program related captioning in a single language. Therefore, the closed captioning system allows for additional "channels" of program related information to be included in the Line 21 data stream. In addition, multiple channels of non-program related information is possible. These channels are referred to in this way:

FIELD 1

FIELD 2

Decoded or "open" captions may be presented in a number of ways. In addition to various character formats such as upper/lower case, italic, and underline, the characters may "Pop-On" the screen, appear to "Paint-On" from left to right, or continuously "Roll-Up" from the bottom of the screen. Captions may appear in different colors as well. The way in which captions are presented, as well their channel assignment, is determined by a set of "control codes" which are transmitted along with the alphanumeric characters. They require both bytes of a given line (a "byte-pair"), and are usually transmitted twice in successive frames (called "byte-pair doubling"). The actual alphanumeric data conforms in most respects to the seven-bit ASCII format, except for some special characters (such as a musical note to indicate musical accompaniment) which require two bytes and are also subject to byte-pair doubling.

It should be noted that transmitting control codes is done at the expense of the   alphanumeric data. "Roll-Up" captions make the most efficient use of control codes and two language captioning on a single field is feasible. On the other hand, the single- field data bandwidth may be insufficient for precisely timed "Pop-On" captions in one language! CLOSED CAPTIONING SYSTEMS:

There are two main divisions into which closed captioning systems may be divided: On-Line (or Real-Time) systems and Off-Line systems. Let us examine each one in turn...

On-Line Captioning:

Certainly the most demanding and exciting of all captioning systems is the stenographic-based  on-line system. This requires a "steno-captioner" using a  courtroom stenograph machine which produces a phonetic representation of the spoken word. Unlike even the fastest typist, a stenographer is able to keep up with the 200+ words per minute of spoken dialogue. Modern steno machines provide a serial data feed, typically RS-232, that is fed to a computer. The heart of the stenographic-based caption system is the computer-aided-transcription (CAT) software running on the computer. Here, the "steno" data is translated into its English (or other language) equivalent. In some cases, the steno-captioner may be assisted by a coordinator who sits at the computer keyboard and handles such tasks as queuing of pre-scripted stories. The coordinator may also view the decoded "open captions" on a TV monitor and make changes to the caption style and placement as necessary.

The computer now provides the ASCII equivalent of the spoken word plus the appropriate control codes (via a second RS-232 port) to a closed caption encoder such as the LINK PCE-845. The live baseband video feed is also connected to the closed caption encoder. The resulting video output is identical to the input except that the closed caption waveform/data has been inserted on Line 21.

An interesting aspect of On-Line captioning is that either of the two serial data links, from the steno keyboard or from the computer, may be handled via modem, a situation known as "remote captioning". A TV station would not necessarily have to have a full-time steno-captioner on-staff but rather could work with a caption service provider on an "as needed" basis.

A second means of handling On-Line captioning is the teleprompter-based solution. This method is far inferior to the stenographic-based solution in terms of the end result to the viewer, but it is somewhat popular nonetheless with broadcasters. In this situation, pre-scripted stories are fed from the newsroom automation system to both the teleprompter and to the closed caption encoder. This data typically does not contain closed caption control codes but rather relies on the encoder itself to insert the appropriate codes whenever an ASCII carriage return is received. The popularity of this type of system is based on the fact that neither a steno-captioner nor a computer with (relatively expensive) CAT software is needed. In many cases, the teleprompter system may already be in place and closed captioning may be added for only the cost of the encoder. The disadvantage of this type of system is that only the pre-scripted information will be captioned. Live reports or other deviation from the teleprompter script will not be supported.Off-Line Captioning:

An important part of today's post-production work, Off-Line Captioning combines precision and creativity. Indeed, high quality off-line software can offer its user complete control over the closed caption data, adding another dimension to the captioned program material. A typical off-line captioning system would work in this way:

The off-line caption service provider typically works with a duplicate of the master videotape of the program to be captioned. In some cases, the script of the program material may be provided on a computer diskette along with the videotape. This script is loaded into a computer that is running the captioning software and the text is   "formatted" to add the desired closed caption control codes. (The program's producer or other creative consultant may take part in this process.)

As mentioned above, these codes determine the location at which the captions will appear on the screen, their color, and style (Pop-On, Paint-On, or Roll-Up).

After the text file is formatted, the videotape is played back and a time code reader, usually installed in one of the computer's expansion slots, is used to assign each caption to a particular time code. Now, everything is ready for the dubbing process by which the closed caption master videotape is produced...

To produce the closed caption master videotape, the original master tape is played back with the video output feeding the closed caption encoder and the time code from the master tape feeding the time code reader card in the computer. When the incoming time code matches that of a particular caption, that caption is transmitted via RS-232 to the closed caption encoder. In this way, the captions can be matched to the video with frame accuracy. The closed caption encoder output goes to a second videotape machine which records the new master tape with closed caption data on Line 21. 

EVALUATING CLOSED CAPTION EQUIPMENT:

In both On-Line and Off-Line captioning, the video path through the closed caption encoder is critical. In the case of On-Line captioning, the encoder's output is sent to the transmitter for broadcast. In the case of Off-Line captioning, the encoder's output is the source for the closed caption master videotape (which may eventually be broadcast). Fundamental video performance specifications such as differential phase and gain, signal-to-noise, chrominance-to-luminance delay, and frequency response should all   be carefully examined. Other points to consider: Does the encoder strip and re-insert incoming caption data? This introduces a 1 frame delay and may also introduce data errors if the incoming data is noisy or marginal in some other way. However, this is important if the user wishes to insert text information (T1 or T2) into incoming caption data. Does the encoder "frame and double" the command codes? If so, can this feature be turned off? Does the encoder accommodate remote captioning with an internal or external modem? Can the encoder be used in teleprompter-based solutions, inserting the appropriate command codes when it receives an ASCII carriage return? Will it automatically erase captions after a particular amount of time? Is this time period adjustable? In any case, choosing the right encoder will depend upon the application in which it will be used.

The closed caption decoder's video performance is critical as well, particularly if an open captioned broadcast or videotape production is required. In addition to clean video performance, the decoder should be able to accurately handle non-timebase corrected video from a tape machine and noisy or otherwise marginal signals.

It is important to note that many types of equipment in a television studio may blank Line 21, effectively deleting the closed caption data there. Therefore, it is a good idea to monitor any closed caption video that passes through a video processing amplifier, video switcher, or timebase corrector. While relatively inexpensive set-top decoders (and new television receivers) are capable of monitoring the closed caption data, only a top-quality decoder will offer the performance necessary to provide open captions for broadcast.In Conclusion...

Closed captioning is a new and exciting field that offers many opportunities for both the television broadcaster and the video producer. Hopefully, this introduction will allow the reader to make intelligent decisions as they choose a caption service provider, purchase equipment, or offer advise to those who will make these choices. The importance of closed captioning can only increase as television receivers without decoders are replaced by new decoder-equipped TVs.A Note About LINK Closed Caption Equipment:

The professionals at Link Electronics have been providing the broadcast industry with high-performance video equipment since the early 1970's. Our equipment for video captioning is of the same superb quality as our more well-known video timing and distribution gear. Here's what we have to offer:

Link is currently shipping outstanding closed caption decoders for the video professional: the IEC-786 & IEC-788. Several closed caption service providers have evaluated the encoders and decoders. One enthusiastic reviewer wrote: "Congratulations on a very nice design. We are very impressed by the product's video handling capability, compact size, and overall excellent performance." We believe you will be impressed as well.

Released in early 1994, the Link PCE-845 closed caption encoder has been very popular as well. Like the decoder before it, the PCE-845 has excellent video handling capability. Evaluated by a major PBS affiliate, the Link encoder's performance was praised. This particular user noted: "Very few devices we operate have video S/N ratios above 65db." (They had measured the PCE-845 at 68.5db RMS.) In addition, differential phase and gain are less than 0.1 degrees and 0.1 percent respectively. Perfectly suited for both On-Line and Off-Line captioning, the PCE-845 is supported by a number of software vendors. (Call Link for a list of compatible software packages.) All operating modes can be controlled via RS-232 with the current mode of operation indicated by front panel LEDs. The current line-up of closed caption encoders are; PCE-845, PDR-885, PDE-890, and PDP-886.

To insert closed caption data into a component digital, SMPTE 259M, bitstream, the Link PDE-890 or PDP-886 is to be used. These two encoders feature digital input and output with 10-Bit resolution. The closed caption waveform is digitally synthesized and is perfect for captioning SDI, D5 and digital BETA Cam formats. A Built-in closed capti on decoder with composite analog output provides an easy means of monitoring the captioned video.

We at Link Electronics welcome your questions and comments. Need a solution to a system timing, distribution, encoding, decoding, or transcoding problem? Think LINK!

REFERENCES:

1. Report No. E-8302, Television Captioning for the Deaf, Phase II - Final Report,            Project Director: David Sillman, Public Broadcasting Service, May 31, 1983.

2. Television Decoder Circuitry Act of 1989/90, 101st Congress, (House Resolution       4267, Senate Resolution 1974). Senate sponsors: Harkin, McCain, Inouye, and             Simon.

3. Telecaption Demonstration Program, Daniel Fapp, Sears Marketing Public                 Relations, January, 1980

4. EIA-608, Recommended Practice for Line 21 Data Service, Electronic Industries Association, Draft of June 17, 1993.

5. Emerging Business Opportunities Through Captioning In The Nineties, Carlos W. Suarez, Cheetah Systems, Inc., NAB 1993 Broadcast Engineering Conference Proceedings.

6. Closed Captioning Extended Data Services: Broadcaster Technical Implications, Paul Kempter, Nielsen Media Research, NAB 1994 Broadcast Engineering Conference Proceedings.

7. FCC Report and Order FCC 91-119 as modified by FCC Memorandum, Opinion and Order FCC 92-157, Closed Caption Decoder Requirements, Federal Communications Commission, March 23, 1992.

The information of this page was prepared by, and reproduced here, courtesy of:
LINK Electronics, Inc.
2137 Rust Ave.
Cape Girardeau, Missouri 63703

Automated Data Systems of Wisconsin, Inc. is an authorized LINK Reseller.