FCC caption requirements
The FCC’s captioning rule for US television is 47 CFR §79.1, "Closed captioning of televised video programming." It obligates video programming distributors (broadcast stations, cable and satellite providers, and others who deliver programming to homes) to pass captions through, and obligates video programmers to caption new programming. The rule then names four specific quality standards captions must meet, each with its own regulatory definition.
Those four names, accuracy, synchronicity, completeness and placement, are worth knowing individually, because "the FCC caption rules" usually means a failure against one specific one of the four, not the rule as a whole.
Check caption timing and structure for free →
The four quality standards, defined
§79.1(j) states it plainly: "Captioning shall be accurate, synchronous, complete, and appropriately placed as those terms are defined herein." Each term gets its own definition in the rule.
Accuracy: captions must match the spoken words, or song lyrics carried on the audio track, in their original language and in the order spoken, with correct spelling, punctuation, capitalization, tense and numbers, and without paraphrasing except where time constraints require it. The same paragraph, §79.1(j)(2)(i), requires more than the words: captions must also convey nonverbal information — who is speaking, the existence of music, sound effects, audience reaction — "to the greatest extent possible."
Synchronicity: captions must coincide with the corresponding speech and sounds "to the greatest extent possible," beginning when the speech or sound begins and ending approximately when it ends, displayed at a speed a viewer can actually read.
Completeness: captions must run from the beginning to the end of the program, to the fullest extent possible. A section can’t be silently dropped just because it was hard to caption.
Placement: captions must be viewable, must not block other important visual content including character faces and featured on-screen text, must use a legible font size, must not overlap each other, and must not run off the edge of the screen.
Live programming is judged differently
§79.1(j)(3) sets a separate, more forgiving basis for live and near-live programming: overall accuracy and understandability, whether the distributor took feasible measures to minimize caption delay, and the inherent transmission lag involved in captioning something as it airs. A live caption stream that would fail the prerecorded synchronicity standard outright can still be compliant under this separate test.
These four standards were established by the FCC’s 2014 caption-quality order (FCC 14-12), adopted in February 2014.
Who has to caption, and who is exempt
§79.1(d) lists 13 exemption categories. Twelve are self-implementing — a channel can rely on them without filing anything — among them non-English/non-Spanish programming, overnight programming between 2 a.m. and 6 a.m., interstitials and public service announcements of 10 minutes or less, new networks in their first four years, and programming that is primarily non-vocal music. The thirteenth, §79.1(d)(2), is different in kind: an economic-burden exemption the FCC grants only on an individually filed petition, case by case.
Two financial thresholds sit alongside the categorical exemptions. §79.1(d)(11) caps captioning spend: no provider has to spend more than 2% of a channel’s prior-year gross revenue on captioning. §79.1(d)(12) exempts channels with less than $3,000,000 in annual revenue from the obligation to create new captions, though the pass-through duty for captions that already exist still applies.
Online video: the CVAA extension
The captioning duty extends online through 47 CFR §79.4, which implements the Twenty-First Century Communications and Video Accessibility Act (CVAA). The rule is narrower than people often assume: IP-delivered programming must carry captions when it was "published or exhibited on television in the United States with captions." The video programming owner has to send the file with captions included; the distributor or online provider then has to render or pass them through.
Content that was never shown on US television with captions is outside this specific mandate. That distinction is why the same show can be captioned when a network airs it and uncaptioned on some catalog services, depending on whether it ever had a captioned US broadcast run.
What a file-level QC pass can and can’t tell you
Two of the four standards are file-level problems: synchronicity is a timing check, and completeness is a coverage check. Both are things a tool can verify automatically from the caption file and its timestamps, without touching the video. The other two are not: accuracy means comparing caption text against what was actually said, and checking that the nonverbal side the rule also demands — speaker identification, music, sound effects, audience reaction — made it into the captions; placement means checking a caption’s position against faces and on-screen graphics in the picture. Both need a human, or a system that watches the video and audio, not just the caption file.
CueForge’s QC checks the file-level side: cue timing and sync, reading speed, gaps, overlaps, structural integrity. It does not compare caption text to the audio track and does not check caption placement against the picture; that’s accuracy and placement, and they stay a human-review step. For delivery, CueForge also reads and writes the broadcast formats these workflows actually use, SCC and EBU-STL, behind the Pro tier; running the free checks first, then exporting to the format your distributor expects, is the same "check free, deliver with Pro" pattern as the Netflix guide.
For the US web-accessibility side rather than broadcast, see ADA closed captioning requirements. For the EU equivalent, see EAA subtitle requirements.
Frequently asked questions
What are the FCC’s four caption quality standards?
Accuracy, synchronicity, completeness and placement, as named and defined in 47 CFR §79.1(j). Accuracy means matching the spoken words and conveying nonverbal information like speaker identity, music, sound effects and audience reaction; synchronicity means coinciding with the audio; completeness means running the full length of the program; placement means not blocking important picture content and staying legible and non-overlapping.
Are live captions held to the same standard as prerecorded ones?
No. §79.1(j)(3) judges live and near-live programming on overall accuracy and understandability, feasible delay-minimization measures, and the inherent transmission lag involved, rather than the stricter prerecorded synchronicity standard.
Is all TV programming required to have captions?
§79.1(d) lists 13 exemption categories, twelve of them self-implementing — including non-English/non-Spanish programming, overnight programming, short interstitials and PSAs, new networks in their first four years, a 2%-of-revenue spending cap and a $3,000,000 annual-revenue exemption from creating new captions. The thirteenth, §79.1(d)(2), is an economic-burden exemption granted only on an individually filed petition.
Do online-only videos need FCC-compliant captions?
Only if they were published or exhibited on US television with captions. That is the trigger for the CVAA-based obligation in 47 CFR §79.4. Content that never aired captioned on US television is outside that specific mandate.
Can a QC tool check FCC caption accuracy and placement automatically?
Not fully. Synchronicity and completeness are file-level checks a tool can verify from the caption file’s own timestamps. Accuracy (matching the audio, including nonverbal information like speaker IDs and sound effects) and placement (avoiding faces and on-screen graphics) both require comparing the captions to the actual picture and audio, which stays a human-review step.
General information, not legal advice — confirm obligations with counsel.
Browse every subtitle converter · Edit subtitles in the browser · Open the CueForge studio