Why Burned-In Text Cannot Be Turned Off

Text can be delivered alongside a video as a separate file or drawn permanently into the frames. The second approach solves distribution problems and creates several others.

Permanent text becomes picture

Once rendered into frames, captions are indistinguishable from any other part of the image. The letters are pixels, not characters.

Nothing downstream can identify them as text. They cannot be extracted, edited, translated or hidden without processing the video itself.

The one advantage is certainty. The text appears wherever the video plays, regardless of what the player supports.

Viewer control disappears

Separate caption files can be resized, restyled, repositioned or switched off. Viewers with low vision often rely on enlarging them substantially.

Burned-in text offers none of that. Its size and contrast are fixed at whatever the producer chose on their own screen.

It also cannot be replaced with another language, so a burned-in version serves exactly one audience.

Compression treats it badly

Hard-edged lettering is expensive to encode and prone to ringing artifacts. Text burned into a frame competes for the same bitrate as the footage.

Every re-encoding degrades it further. Text delivered separately is drawn by the player at full display resolution and stays crisp regardless of the video's condition.

Scaling makes this worse. A video downscaled for a small screen shrinks its burned-in text to the point of illegibility, while separate captions stay readable.

Machines cannot read it

Search engines and platform indexing systems read caption files to understand video content. Burned-in text is invisible to them.

Assistive technologies that convert text to speech or to a braille display likewise depend on actual text. Pixels are inaccessible to them.

Accessibility requirements generally contemplate captions the viewer can control, so burned-in text may not satisfy an obligation even when the words are present. Rules vary by context and change.

There are situations where it is the right choice

Some distribution paths strip separate caption files or offer no way to enable them. Text that must survive republishing may have to be burned in.

Feeds where video autoplays silently in a scrolling interface also favor permanent text, since nothing prompts the viewer to enable anything.

The reasonable compromise is producing a separate caption file as the master and burning in a version only where the delivery path requires it. The editable original still exists.