iCouldPlay VTT emphasis: pair nested tags without changing the sentence

Emphasis markup may affect appearance but does not decide a sentence’s importance. Add nested spans to an original short cue, use existing iCouldPlay loading/playback controls to check wording, then record observed styling. Edit externally and keep a plain fallback without assuming browser-equivalent rendering or an in-app editor.

1. Fix wording and timing for one clip

Choose an owned, playable clip of at least ten seconds without burned-in exercise wording. Record source, file identity and caption controls; prepare the plain wording first. State that nesting is the tested variable, holding font, size, orientation and video fixed. Without an available loading control, complete external inspection and mark playback unperformed; a file listing cannot establish rendering. Save new files separately from the original video and everyday captions.

2. Save three complete original cues

Save the complete sample as actual UTF-8 plain-text VTT. Cue one is plain; cue two requests bold Keep, bold-italic inner and bold wording; cue three underlines only inner. Visible wording is identical with separate intervals. This is an original test sentence, not film dialogue. Compare words and punctuation before tags. Do not insert a web-document wrapper around timings or replace closing-tag slashes with backslashes.

WEBVTT

00:00:01.000 --> 00:00:03.000
Keep inner wording.

00:00:04.000 --> 00:00:06.000
<b>Keep <i>inner</i> wording.</b>

00:00:07.000 --> 00:00:09.000
Keep <u>inner</u> wording.

3. Match inner and outer spans

Use the b/i/u spans described by the W3C WebVTT text-markup specification. On paper, pair cue two’s openings and closings: close the inner italics before the outer bold, avoiding crossed boundaries. Mark inner as inside both spans and the remaining words differently; independently check cue three’s single-word underline. Format roles and actual font appearance need separate records. Tags alone do not pass iCouldPlay rendering or establish an author’s spoken stress.

4. Check wording before appearance in playback

Select the exercise copy and play continuously from zero to beyond nine seconds. Transcribe each full sentence before describing word-level appearance. Resume after pausing to check cue endings rather than mistaking a paused frame for persistence. Report no observed styling difference honestly. Intact text and unverified styling are separate outcomes. Keep, inner, wording and the period remain required; an emphasized inner word does not excuse omitting surrounding text or punctuation.

5. Make a timing-matched plain fallback

Save a plain fallback by removing only tags from cues two and three, retaining all words, spaces and timings. Verify selected filename/source when switching; identical wording can hide a mistaken selection. Keep video, display mode and playback path fixed. Switching engine creates a new condition rather than proving a unique cause. The fallback preserves readable content without erasing the styled original or claiming that the cleaned copy was the author’s initial file.

6. Separate markup faults from interpretation faults

Write a crossed-closing teaching error on paper, labeled for source inspection rather than routine playback. Identify crossed spans and repair inside-out without also changing timing or wording. Valid bold does not prove shouting, and underlining does not prove a hyperlink. Record an attributed source explanation if one exists; otherwise retain appearance-only candidates. Interpretations require their own evidence rather than inferring emotion or events from styling.

7. Troubleshoot exposed tags, missing words and identity

For literal angle-bracket text, check actual VTT format, spelling and escaping rather than hiding it with a smaller font. Missing inner calls for checking accidental text deletion before blaming fonts. One absent cue calls for checking boundaries and selection. Do not invent unseen app logs or clear the media library. Change one identified condition and replay from before that cue; a multi-file change that happens to work cannot establish one unisolated cause.

8. Reopen and deliver span and observation records

Leave playback, reopen the same clip, reselect the exercise and replay all three sentences. Keep the span map, complete transcription, intervals, observed appearance, fallback and unknowns. Valid nesting must not become universal styling support, nor visual emphasis proof of an author’s or character’s intention. Replay the plain fallback before finishing and preserve originals. After changing size, source or rendering path, append another conditional record rather than rewriting an earlier observation.

Related tutorials and sources

VTT class spans, VTT style blocks.

Checked on November 13, 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.