What are subtitles and closed captions?

Subtitles and closed captions are timed text shown on screen while a video plays. Subtitles carry the spoken words, often translated into another language, while closed captions also describe sounds such as music or an alarm, and the viewer can switch them on or off.

Reviewed by Clark Quinn, Chief Learning Strategist Updated October 28, 2026

Example

A manufacturing company’s L&D team publishes a short safety video for shift workers. Many of them watch it on a phone in the break room with the sound off, and some are deaf or hard of hearing. The team adds closed captions that include the spoken instructions plus cues like “[reversing alarm]”, so the warning sound in the video still reaches anyone watching without audio.

The same video then goes to a plant in Mexico. Instead of recording new narration, the team adds a Spanish subtitle track. Workers at both sites watch the same file and pick the text track they need in the player.

How it works

Subtitles and captions are made of short text segments, often called cues. Each cue has a start time, an end time and one or two lines of text. The video player shows each cue at its start time and hides it at its end time, which keeps the text in sync with the audio.

Creating the text

There are three common ways to produce captions:

• From a script: if the video was built from a written script, as narrated and AI-generated videos usually are, the script is split into cues and timed to the audio.
• Speech recognition: automatic speech recognition (ASR) transcribes the audio and times it. This is fast, but a person needs to check names, jargon and acronyms.
• Manual transcription: a person types and times every caption, often through a professional captioning service for high-stakes content.

Caption files

Closed captions and subtitles usually live in a separate text file that travels with the video. The two formats most players and learning platforms accept are:

• SRT (SubRip): a simple list of numbered cues, each with its timestamps and text.
• WebVTT (.vtt): the format built for HTML5 video players, with more options for styling and positioning text.

Because the file is separate from the video, one video can carry several tracks, one per language, and the viewer picks which one to show.

Open vs closed captions

Closed captions can be turned on and off because they sit on a separate track. Open captions, also called burned-in captions, are rendered into the video image. Open captions appear in any player or social feed, but nobody can switch them off, restyle them or swap the language without exporting the video again.

Subtitles vs closed captions

The terms overlap, and in British English “subtitles” often covers both. In US usage, which most video platforms follow, the difference comes down to who the text is for.

• Subtitles assume the viewer can hear the audio but cannot follow the language, or has the sound off. They show the dialogue and little else. Translated subtitles are one of the simplest tools for video localization: they keep the original audio, while AI dubbing replaces it with a voice track in the new language.
• Closed captions assume the viewer cannot hear the audio. They add speaker names and non-speech information, such as [phone ringing] or [laughter], because those sounds carry meaning.
• SDH (subtitles for the deaf and hard of hearing) are subtitles written to caption standards, with speaker names and sound cues. They are common on platforms that only support subtitle tracks.

Best practices for training video

• Keep each caption to one or two lines, and break lines between phrases rather than in the middle of one.
• Leave each caption on screen long enough to read. Fast narration may need slightly shortened caption text.
• Check automatic captions for product names, acronyms and numbers before publishing.
• Describe sounds that matter to the lesson, such as an alarm in a safety video, and leave out background music that does not.
• Make sure captions do not cover on-screen text or the part of a screen recording the learner needs to see.
• Use the same terms in captions, on-screen text and course materials.

Why it matters for L&D teams

• Sound-off viewing. Learners often watch on phones in shared spaces, on shop floors or in open offices. Captions let them follow the lesson without audio.
• Accessibility. Captions make video usable for people who are deaf or hard of hearing, and many organizations’ accessibility policies require them on training content.
• Comprehension. Reading along helps learners catch technical terms, names and numbers, and helps non-native speakers follow the narration.
• Faster localization. A translated subtitle track lets a team ship one video to several regions without new narration, often as a first step before full video translation.