Captions and subtitles

Both are text on the frame that shows what is being said. Strictly, subtitles carry the dialogue and assume the viewer can hear the rest, while captions also describe the sounds, the music and who is speaking, for a viewer who cannot hear at all. In screen recording the distinction has mostly gone: what people want is the words on the frame, and either name will get them.

Why a screen recording wants them

A recording is opened where its sound cannot be played. On a phone, in an office, in a feed that autoplays muted. The viewer sees the pointer move and hears nothing, and decides in those first seconds whether it is worth the headphones. Captions are what carries the narration through that decision.

They also make the recording searchable. A transcript beside a video is text a search engine can read, and a caption file is what lets a player offer the words in another language.

Burned in or a file

Burned in, the captions are part of the picture: drawn onto every frame, styled however the maker chose, and always there. This is what a feed, a chat, a Slack message or an email needs, because none of them offers a subtitle button.

As a file, an SRT or a VTT, the captions travel beside the video and the player draws them. The viewer can turn them off, a platform can index them, and the same file can be translated. YouTube, Vimeo and a share page can all use one.

Where the words come from

Typing them is the slow way. A video editor with transcription is faster, and an upload to a captioning service is faster still, at the cost of sending an unreleased product to somebody's server. A recorder that transcribes on the Mac as the recording opens is the version with no waiting and no upload, which is what Prequel's captions do: the transcript appears in the editor as text to type into, the captions are drawn from it, and a share link offers the same words as subtitles.