NCI's Live Captioning

BACKGROUND:

In 1982, the National Captioning Institute (NCI) developed the real-time captioning service to give captioned TV viewers access to live programming. With the world's most sophisticated real-time captioning facilities and the most highly trained real-time captioning professionals, NCI provides over 50,000 hours of live captioning every year for the television industry, corporations and government agencies.

WHAT IS LIVE CAPTIONING?

Live Captioning is provided at the time of program origination. NCI provides three kinds of live captioning services:

Real-time Captioning: Roll-up captions are created and displayed at the time a program is on the air.

Live-display Captioning: Roll-up captions are prepared in advance but transmitted at the time a program is on the air.

Live Encoding: Pop-on captions are prepared in advance but transmitted at the time a program is on the air using Automatic Live Encoding (ALE).

HOW IS REAL-TIME CAPTIONING CREATED?

Real-time captioners use a computerized system based on the stenographic shorthand used by court reporters. A real-time captioner, someone who has been trained to transcribe speech to text using a steno machine, listens to a program's dialogue as it is being broadcast and enters the words phonetically in stenographic shorthand code. The steno machine is connected to a computer containing software that translates stenographic shorthand into words using standard spellings and then converts them to a caption format. The caption data is transmitted instantaneously via modem to the master control at the network. There, the caption data is inserted into Line 21 of the television signal through a smart encoder. Since the real-time captioner cannot transcribe a word until it is spoken, real-time captions always lag slightly behind the audio, generally by about two to three seconds. Real-time captioners may be located at NCI's Vienna, Virginia headquarters or at any of NCI's 40 remote locations in 11 states.

Since real-time captions are being created and displayed while the program is being broadcast, there is no opportunity to proofread captions or to replay difficult to understand audio. As a result, errors do occur, usually in the form of incorrect, though phonetically similar, words (such as "row place" instead of "replace"). NCI continuously assesses each real-time captioner's work so that accuracy rates of 98 percent or better can be maintained.

WHAT SKILLS ARE NEEDED FOR THIS DEMANDING WORK?

NCI recruits court reporters and retrains them to become real-time captioners. It can take up to a year for a court reporter to develop the speed, accuracy and skills necessary to be a real-time captioner who is capable of creating captions from the spoken word at over 225 words per minute.