Choose according to the task
An owned video can deliver captions separately, inside a supported media package, or visibly composited into the picture. Each approach serves a different workflow. Begin with the audience's need, the destination's documented capability and the likelihood of later corrections.
W3C WAI distinguishes closed captions that viewers can show or hide from open captions that remain visible. That viewing choice is important, but packaging also affects editing, translation and reuse. Keep the approved clean picture and editable caption source where the production requires future versions.
Test the relevant destination behaviour
A sidecar is useful only when the destination associates it with the right media and exposes it appropriately. An embedded track requires support for its actual format, not merely the enclosing extension. Burned-in text becomes part of the picture and needs visual review at every intended delivery size.
Imagine a hypothetical training video delivered to a platform that accepts a caption sidecar and to a fixed display with no text-track selection. The production can maintain one editable caption source while creating two deliberate derivatives. It should not assume a single file will serve both destinations identically.
Keep the editorial source separate
Store caption text, timing and necessary speaker or sound information in a version associated with the approved picture cut. Correct the source first, then regenerate the required delivery forms. Editing only a burned-caption copy leaves the next export vulnerable to restoring the old error.
Plan language versions as distinct assets. An audio-language change may need another text version and timing review. A track label should explain what the viewer receives; a generic subtitles label may not distinguish translation from captions carrying meaningful sound information.
Accept each delivery form
For selectable captions, confirm discovery, selection, text and timing. For open captions, inspect readability during motion and through the smallest intended encoding. Review overlay collisions with graphics and controls. A successful upload is only one preparation stage.
At handoff, identify the clean master, editable caption source and derived packages. Record the destination tested and any support limitation. This is a production decision method informed by W3C WAI guidance, not a claim that IPTV Europe or every player offers each packaging option.
Questions answered
Can burned-in captions be switched off by the viewer?
They are part of the picture. Retain a clean master if a version without them may be needed.
Does embedding a caption track guarantee destination support?
No. Check the specific text format and destination behaviour.
Sources & further reading
Official references for the subjects discussed. Service terms and device software can change.