iCouldPlay subtitle origins: distinguish embedded tracks from external files

Two entries labelled Chinese may have different origins and edition matches. Identify embedded tracks and external files, then verify them with recognisable text. Unlike a burned-in/selectable comparison, this focuses on origins of selectable captions without extracting embedded streams or modifying the container.

Choose one video and permitted captions

Use owned media and permitted captions. If no embedded track exists, record that and use an authorised sample that has one for that part; loading a sidecar does not embed it. A separate self-written sidecar prompt can serve as a control while preserving production captions. Record video, duration, edition and acquisition methods without uploading to unfamiliar extraction services. Make an origin map, not a complete subtitle-format remake.

Copy labels from the visible list

Open the same video and copy caption names, languages and any embedded/external markers actually shown. Without origin markers, label the evidence insufficient rather than assuming first means embedded. A file you just loaded has an observable origin; pre-existing entries still need checking. Record the current selection/off state and change one item at a time, since refreshed list indices are not permanent identities.

Record sidecar identity separately

For sidecars, note name, format, modification date and a distinctive short payload, using private location aliases. Same names or dates do not establish identical content. A prompt such as Keep this sample separate distinguishes a self-made control from dialogue within a known interval. Label it as practice text, not a translation. iCouldPlay loads/displays captions while external editing and version management remain separate.

Verify the first selection by content

Select one entry and play from before a cue, recording actual wording, approximate file time and visual anchors. A Chinese label showing commentary remains commentary, not automatically primary subtitles. Make available style adjustments only as needed, preserving originals. Text that persists after captions are disabled belongs to another layer and cannot validate the selected track. Attribute each observation to one specific entry.

Compare another origin at the same position

Select another origin and replay from the same earlier position, recording wording, timing, language and segmentation. Agreement at one anchor does not establish identical files or timelines. For absent text, check selection, actual sidecar loading and whether that interval contains cues before declaring track damage. Keep media and engine fixed so the map still identifies one object.

Resolve ambiguous labels with traceable notes

For identical labels, distinguish the newly loaded sample from the original list entry plus observed text, documenting that session’s index only. Do not rewrite all dialogue or assume an embedded-track rename button. Rename an owned copy externally only if necessary, reload and verify it. Keep uncertain origins unconfirmed rather than inventing a convincing label.

Check selection again after reopening

Reopen the same video, inspect current selection and replay the anchor to establish the active origin. Continued sidecar availability depends on actual storage/loading, not a universal persistence promise. Record moved files separately rather than calling a broken sidecar path a vanished embedded stream. Restore normal captions and keep practice names separate without clearing the whole subtitle directory.

Check the current edition before timing repairs

If only one origin matches dialogue, verify media edition, opening and subtitle notes before the existing beginning/middle/end timing check. Different editions need not be repaired by a global shift. Different wording may be another translation or commentary. Record specific differences, preserve originals and separate identity, text and timing questions rather than overwriting the only file.

Finish the caption-origin map

Keep media identity, labels, origin evidence, selected/reopened outcomes and unknowns. Another reader should locate the observed cue, not merely two Chinese list entries. Public embedded/external support comes from the store description, with actual choices verified in the installed version. No extraction, upload or exhaustive device-format testing was performed. Preserve media and production captions and remap identities after edition changes.

Related tutorials and sources

Engine checks, Original and processed copies.

Checked on November 8, 2026 against the official iCouldPlay App Store listing and Apple public lookup: version 1.1.1, requiring iOS / iPadOS 15.0 or later. The public 1.1.1 release notes list Kodi, OpenList / AList and S3-compatible storage, with improvements to source switching and captions after seeking. The tasks use this public scope and existing playback capabilities. Check controls in your installed version. Self-made samples, manual records and external file preparation are workflows, not newly promised automatic features. No physical-device or private-server test is claimed.