Captioning Style Guides: 8 Standards Worth Bookmarking

Style guides settle the arguments. How many characters per line. Whether to caption music. Whether [music] is bracketed or italicized. Whether a speaker's name goes on a new line or inline.

Most captioning teams write their own internal style notes, but those notes almost always crib from a handful of public reference guides. If you caption anything for a paying audience — a streaming platform, a broadcaster, a US federal contract, a school district — you're being measured against one of these documents whether you've read it or not. So read it.

Here's a curated list of eight public caption and subtitle style guides worth keeping in a tab. I've grouped them by where they apply — streamers, broadcasters, accessibility bodies, and industry vendors — so you can find the one that actually governs your work.

Why caption style guides exist

Captions look simple from the outside. They're not. Reading speed varies by audience and by language. Two-line captions may be 42 characters per line on Netflix English and 37 on the BBC. Music cues are bracketed on one platform and italicized on another. A delivery that fails QC on a streaming platform may be perfectly compliant for a school district, and vice versa.

The point of a style guide isn't pedantry. It's making sure a deaf or hard-of-hearing viewer in row B of a multiplex sees the same caption a viewer streaming the same title at home does — accurate, readable, and timed to the dialogue. Style guides codify what "accurate, readable, timed" actually means.

Streaming-platform style guides

If your work ships on Netflix, Amazon Prime, Disney+, Apple TV+, or Hulu, the platform's own guide is the only one that matters at QC.

The Netflix Timed Text Style Guide is the best-known of the bunch and arguably the most influential — many other streamers' guides borrow from it heavily. It defines characters per line, reading speed (17 cps for adult programming, lower for children's), how to handle speaker IDs, music, sound effects, on-screen text, foreign dialogue, and forced narratives. Netflix publishes language-specific guides too — Spanish-LAS, Brazilian Portuguese, Japanese, Arabic, and dozens more — each with its own reading speed and orthographic rules.

If you're working on streaming captions or subtitles for your videos using SRT files, this is the document to anchor on.

Broadcast-network style guides

Broadcast captions are older than streaming captions, and the standards are stricter on placement, timing, and music.

The BBC Subtitle Guidelines (hosted on GitHub, freely browsable) are the touchstone for UK and European broadcast subtitling. They're opinionated about pop-on vs roll-up captions, placement to avoid covering speakers' mouths, and the use of color to differentiate speakers — a practice US streamers have largely abandoned but UK audiences still expect.

In the US, the FCC closed captioning rules govern accuracy, completeness, synchronicity, and placement for most television and TV-distributed content rebroadcast online. The FCC's quality standards are looser than Netflix's on prosody but mandatory on synchronicity (captions cannot lag dialogue) and on placement (captions must not obscure relevant on-screen text). Enforcement is real.

UK broadcasters are governed by OFCOM's Code on Television Access Services, which sets minimum percentages of programming that must be subtitled, signed, and audio-described, plus quality thresholds. If you're delivering to a UK broadcaster, your contract almost certainly references it.

Accessibility-first style guides

Where streaming guides optimize for the editorial experience and broadcast guides for compliance, accessibility guides start with the deaf and hard-of-hearing audience and work outward.

DCMP's Captioning Key (Described and Captioned Media Program, funded by the US Department of Education) is the most thorough public accessibility-led caption guide in existence. It's used by schools, universities, and US federal contractors, and it covers edge cases — overlapping speakers, off-screen sound effects, music with lyrics, untranslated foreign dialogue — that other guides skip.

WCAG 2.2 doesn't tell you how to write a caption. It tells you when one is required. Success Criterion 1.2.2 (Captions, Prerecorded) is what makes captions a Level A web accessibility requirement; 1.2.4 (Captions, Live) escalates that to AA for live content. If you're working through our WCAG captions compliance checklist, you're already living inside WCAG's expectations.

NCAM at WGBH publishes practical captioning standards and tools, including the open-source CCPlayer and Caption Maker projects. NCAM's documentation is the bridge between WCAG's "you must" and DCMP's "here's how" — it explains the rationale for accessibility choices in language a captioner can actually act on.

Vendor and industry style guides

The big captioning vendors publish their own style guides, partly for their staff and partly as marketing assets. The good ones are a sensible blend of the platform and accessibility standards above.

3Play Media's captioning style guide is the most comprehensive of the vendor guides. It pulls streaming-platform conventions, FCC requirements, and DCMP-style accessibility guidance into one reference your captioner can finish reading in an afternoon. It's a strong starting point if you don't yet have a target platform, or if your target is a corporate buyer who doesn't have one either.

What every style guide covers (and what none of them cover)

Across the eight, the consistently covered topics are:

Reading speed — characters per second, slower for children's programming
Line length — characters per line, two-line maximum, balance between the lines
Timing — pop-on vs roll-up, lead/lag relative to audio, minimum gap between captions
Music and sound effects — when to include, how to mark them, when to italicize
Speaker identification — name labels, color coding (UK only), positional cues
Profanity and proper nouns — verbatim vs euphemized, name verification

What no public style guide tells you:

What to do with bad source audio. Style guides assume the audio is clean. When it isn't, you're guessing — and the guides won't help. Clean the audio first, or use a tool that's good at noisy input.
How to handle errors at scale. Style guides cover one caption at a time. They don't tell you how to do verbatim QC across a 90-minute talking-head video efficiently.
What "intelligent verbatim" actually means. For that, see our verbatim vs intelligent transcription guide.

Which one should you actually follow?

In order of which guide governs:

The platform you're delivering to, if there is one. If Netflix says 17 cps, the FCC's looser rules don't save you at QC.
The regulator covering your distribution region. FCC in the US, OFCOM in the UK, ACMA in Australia, CRTC in Canada.
WCAG 2.2 if your distribution is the open web and you have no platform contract.
DCMP or NCAM as your accessibility north star, regardless of platform.
3Play or another vendor guide as a tie-breaker on edge cases the others don't address.

If you're starting from scratch with no platform contract, follow DCMP's Captioning Key and you'll satisfy most other guides by default.

Try it now — it's free

Transcribe your video with Ask Giya

Paste any public link or upload a file and get a clean transcript in minutes. First 3 clips every month are on us — no card required.

Start transcribing No subscription · 8¢/min after free clips

Once you've picked a guide, the rest is process — generate the transcript, apply the style rules, time the captions, QC against the guide's checklist. The hard part is choosing which one. Now you have.