Skip to main content

What is a caption file? A beginner's guide with examples

· 5 min read
Recap Innovations
AI-Powered Video Accessibility Solutions

A caption file is a text document that pairs written words with specific moments in a video. Each line of text includes a start time and end time so the words appear on screen in sync with the audio. Caption files do not contain the video itself; they are uploaded alongside the video (or embedded into it) so viewers can read what is being said.

What does a caption look like?​

On screen, captions appear as text overlaid at the bottom (or top) of the video, synchronized with the spoken audio:

┌─────────────────────────────────────────┐
│                                         │
│         [Video content plays here]      │
│                                         │
│  ┌───────────────────────────────────┐  │
│  │ Welcome to today's training.      │  │
│  └───────────────────────────────────┘  │
└─────────────────────────────────────────┘

In the caption file itself, the same words appear with timestamps. Here is a simple SRT (SubRip) example:

1
00:00:01,000 --> 00:00:04,500
Welcome to today's training.

2
00:00:04,500 --> 00:00:08,200
In this section, we'll cover safety protocols.

Each caption block has:

  1. A sequence number (1, 2, 3...)
  2. A time range showing when the text appears and disappears
  3. The caption text (one or more lines)

Closed captions vs open captions vs subtitles​

These terms are often confused:

TermWhat it means
Closed captionsText in a separate file the viewer can turn on or off; includes speech and often sound effects like [door slams]
Open captions (burned-in)Text permanently embedded in the video; always visible, cannot be turned off
SubtitlesText translating or transcribing dialogue; traditionally assumed hearing audience, though the line with captions has blurred

For accessibility compliance (ADA, WCAG, Section 508), closed captions that viewers can enable are the standard requirement for most web video.

Common caption file formats​

SRT (.srt)​

The most widely supported format. Works with YouTube, Vimeo, most LMS platforms, and social media.

Best for: Web video, e-learning, general compatibility

WebVTT (.vtt)​

The web standard for HTML5 video players. Supports basic styling and positioning.

Best for: Web players, Canvas, custom HTML5 embeds

Example VTT block:

WEBVTT

00:00:01.000 --> 00:00:04.500
Welcome to today's training.

SCC (.scc)​

Scenarist Closed Caption format used in North American broadcast television. Encodes CEA-608/708 data for professional broadcast workflows.

Best for: Broadcast TV, cable, FCC-regulated distribution

SCC files look nothing like plain text SRT files; they contain hexadecimal-encoded caption data. If you are working with broadcast or cable distribution, see our guide on CEA-608 vs CEA-708 and the complete caption formats guide.

How caption files work with video​

  1. You create or generate a caption file (manually, via AI transcription, or through a captioning service)
  2. You upload the caption file to your video platform alongside the video
  3. The platform reads the timestamps and displays the matching text during playback
  4. Viewers toggle captions on or off (for closed captions)

Most platforms accept SRT or VTT. You do not need a separate video file for each language; you upload different caption files for different languages on the same video.

Should captions be italicized?​

Sometimes. Professional captioning style guides (including DCMP guidelines) use italics for:

  • Off-screen speaker attribution when the speaker is not visible
  • Foreign words or phrases spoken within otherwise English dialogue
  • Voice-over or narration distinct from on-screen speakers (context-dependent)

Not every platform renders italic styling from caption files. SRT has limited styling support; VTT and TTML support more formatting options. For web video, plain text captions without italics are acceptable when the platform cannot render styling, but professional captioners use italics where guidelines call for it.

Recap's captions follow professional standards including descriptive sound cues and proper formatting for accessibility.

How to create a caption file​

Option 1: Write manually in a text editor, timing each line (slow for long videos)

Option 2: Auto-generate with AI transcription, then review and edit for accuracy. Recap generates caption files automatically from uploaded video.

Option 3: Use a caption converter to convert between formats. Recap offers a free caption format converter for SRT, VTT, DOCX, and TXT.

Option 4: Hire professional captioners for broadcast, legal, or high-stakes content requiring 99%+ accuracy. Recap offers professional caption moderators and transcript editors that review and clean up typos and errors in transcripts and apply speaker tag identification.

What should I put in the caption?​

Captions should include:

  • All spoken dialogue, including speaker changes
  • Sound effects important to understanding ( [applause], [phone rings] )
  • Speaker identification when speakers are not visible (e.g., "NARRATOR:" or "JOHN:")
  • Lyrics if music conveys meaning

Captions should not include:

  • Excessive punctuation or emoji (unless spoken)
  • Commentary or interpretation not in the audio
  • Text that covers essential visual content

Next steps​