In 2015 the FCC promulgated FCC Rule 79.1, including section (j)(2) Captioning quality obligation; standards, which specifically relates to caption quality on television. Section (j)(2) states as follows:
"(2) Captioning quality standards. Closed captioning shall convey the aural content of video programming in the original language (i.e. English or Spanish) to individuals who are deaf and hard of hearing to the same extent that the audio track conveys such content to individuals who are able to hear. Captioning shall be accurate, synchronous, complete, and appropriately placed as those terms are defined herein."
Some key points on the accuracy component are that the captions should match the spoken words without substituting words for proper names and places and without paraphrasing, to the extent possible given time constraints. The captions shall also include correct spelling, punctuation and capitalization. Song lyrics should be provided when they are available on the audio track, and slang and grammatical errors are to mirror the spoken word.
The synchronicity of captioned text may be observed when comparing the time the words are spoken to the time the word text appears on the screen. While a minor time lag is unavoidable due to response time from the captioner hearing and writing the words, the software converting the shorthand words to full text, and the technology transmitting the data to the television station, the entire process from speech to text for realtime captioning is generally approximately 0-5 seconds. Onsite and Internet realtime captions are more timely, however, than television captions, in part due to the built-in delay on broadcast signals and distance the data has to traverse before arriving on the viewer’s screen.
The aspect of completeness addresses the capacity of the realtime captioner to capture the communication throughout the program length.
For pre-scripted programming the placement of captioning on the television screen is controlled by technicians other than realtime captioners; for realtime captioning the placement on the screen of the text is a setting input by the captioner based on the instruction from the television station. However, the live program captioning as it is seen on TV very often is not viewable by the captioner who is creating the realtime text because the captioner is working from a location outside the viewing area of the broadcast and the realtime captions may not be simultaneously viewable on the Internet.
Many of the aspects of the appearance of realtime captions on the television screen are outside the control of the captioner; for example, where the caption text appears relative to graphics or banners placed on the screen, the time from when the caption text arrives at the television station to when it is displayed on the viewer’s television screen, and the accuracy of the captions insofar as missing or garbled captions which may be caused by interference of satellite signals along the signal path. Importantly, many factors adversely impact the quality of captions that are beyond the control of the captioner, the technology, and the television station technicians: a speaker who speaks upwards of 300 words a minute, who does not adequately enunciate his/her speech, who does not speak (or is not amplified) sufficiently loudly, who has a very pronounced accent, or who is speaking at the same time as one or more other individuals.
In addition to television captioning rules, new rules have been developed to require the availability of closed captions for audio-video media on the Internet when such media has been previously broadcast/recorded with captions.
The rule controlling Closed Captioning of Video Programming Delivered Using Internet Protocol, which includes updates effective as January 1, 2016, is 47 C.F.R. Section 79.4.