iCouldPlay visual-information notes: describe tutorial arrows, colours and positions

A tutorial’s “look here” may depend on an arrow, colour or location, so dialogue alone can omit essential information. Use iCouldPlay to inspect your own short clip and write an external visual description, distinct from sound-event captions. This assumes no automatic description, arrow recognition or transcript reading, and does not claim complete video accessibility validation.

Prepare a sample with few visual instructions

Choose an approximately one-minute self-made or expressly describable tutorial using sorted paper cards or fictional onscreen shapes, without private accounts or credentials. Record filename, version, duration, sharing permission and comprehension task. Play normally once and identify information available only visually, without transcribing every frame. Preserve the source and write separately; keep notes private without public-description permission. No video editing, plugin installation or upload to an unknown site is needed. A short sample suffices to test the explanation.

Separate speech, screen wording and visible action

Make three fields: audible explanation, essential onscreen wording, and arrows/object movement. Describe task-relevant content from the latter two rather than repeating music or door-sound captions. In an invented clip, “move it over” is accompanied only by a red arrow from left box to right. Add the object and destination instead of making readers guess “it.” Mark unclear speech unresolved and do not invent motives. Even when combined into a paragraph, preserve which content is original speech and which is your explanation.

Bind positions and directions to a reference

Specify screen-left/right/up/down as viewer-relative, separately describing object turns rather than equating the operator’s left hand with screen-left. For arrows, record start, end and target, not merely a red arrow. Add labels or shapes where colour classifies, such as moving a circular card into the right box marked B. Without labels, use a stable location rather than inventing an official name. Pause to inspect, then resume normally to confirm the intended target instead of mistaking a passing pointer for an instruction.

Save time and content anchors for each description

Use available ordinary seeking to record nearby playback time, starting image and resulting state, without claiming exact frame numbers. An invented interval might begin around twelve seconds when the arrow appears and end when the card rests in the right box. Reading time is not media position. After seeking, wait for stable playback; keep caption timing separate from action and do not call the action complete because text appeared. Recheck these times after source-version changes.

Describe necessary information without excessive detail

Start with one scene overview, then ordered who/what/from/to and essential screen words. Decorative colours and table texture can be omitted, but retain negation, quantity and target labels. Separate inference, and do not treat a recommended arrow as completed action. Without producing audio description, call this an external text draft, not an added audio track or automatically read app feature. Timed description requires separately verified player support and production; do not force it into the densest dialogue here.

Hide the image and test task understanding

Read the draft without the image and check referents for it, here and there and distinguishable before/after states. Add shape or location when several red objects make “the red one” ambiguous. Remembering the image yourself does not prove self-contained wording. An authorised collaborator may identify gaps, but do not claim audience validation without feedback. Revisit for necessary relationships rather than inventing actions. Revise the description copy while retaining source and initial question so each addition has a reason.

Leave the player and return for proofreading

Leave the player, reopen the same file and use time plus content anchors to replay instruction onset, action and result. Check arrow direction, colour labels, numbers and negation against your paraphrase. One screenshot does not replace a movement sequence and remains subject to sharing permission. Allow context for several seconds of seeking uncertainty rather than claiming frame precision. Rematch changed version, source or duration instead of transferring the old draft unconditionally. Restore ordinary viewing position and preferences afterwards.

Finish the draft and state feature boundaries

Keep file identity, time anchors, essential visual information, paraphrase provenance and unknowns. See W3C: Description of Visual Information for methods and player-support conditions, not proof that iCouldPlay offers every description control. Dialogue may depend on visual reference, while more words can obscure the task with irrelevant detail. Retain one improvement and the manual checking record without calling the draft automatic captions, integrated audio description or completed whole-video accessibility validation.

Related tutorials and sources

Sound-event captions, Manual lesson time index.

Checked on November 6, 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.