Find the instruction the audio misses

The audio says “Choose the labeled folder,” but only the picture names Drafts. Accurate captions still leave that destination unspoken.

This HTML video accessibility review separates audio text from visual description before anyone claims a clip is accessible.

Inspect three intervals from a planned 12-second sorting demonstration, comparing caption content with the visual information a viewer also needs. The silent middle interval has no planned audio cue, while its descriptive text identifies moving the sketch into Drafts. A folder-label revision updates the dependent description after an explicit rebuild; a V2 proposal remains unrecorded narration.

Find the instruction the audio misses
Find the instruction the audio misses
Open the full example ↗

Inventory the planned source before adding a track

W3C WAI distinguishes text for speech/meaningful sounds, description of visual information, descriptive transcripts and player support. Adopt those complementary roles rather than assume one track meets every need. W3C WAI: Making Audio and Video Media Accessible

The original storyboard is a fictional sorting demonstration, independent of a verified product interface. Use source version PLAN-V1 and three planned intervals:

Planned interval Audio content Visual content Information gap
0–4 seconds Speaker: “Choose the labeled folder” A sketch and labeled folders are visible Spoken direction does not name the destination
4–8 seconds No planned speech or meaningful sound Sketch moves into folder labeled Drafts Destination/action supplied only visually
8–12 seconds Meaningful completion chime; speaker: “The move is complete” Finished state shown Chime and confirmation must remain in audio-text record

These intervals are production plans, not measured cue timings or observations of a recording. Actual onset, duration, sound clarity and visual legibility remain unknown.

The planned chime signals completion, so its meaning belongs in the caption draft. Confirm the actual sound and timing once a recording exists.

Write captions from the planned audio

For PLAN-V1, the caption-content draft contains the supplied dialogue and meaningful chime:

Interval reference Planned caption content Source reason
0–4 Choose the labeled folder Planned spoken words
4–8 No audio-content cue planned Silent visual action
8–12 [Completion chime] / The move is complete Planned sound and spoken confirmation

The slash separates two content records, not a demand to show both for the whole four seconds. When media exists, determine their actual timing and segmentation from listening and review.

Do not insert Move the sketch into Drafts into the silent interval as fabricated spoken dialogue. Captions can include meaningful sound information; they should not obscure the fact that this version leaves a visual instruction unspoken.

An accurate transcript of the audio alone also lacks Drafts. That is a coverage problem to solve with visual description, not evidence that the audio transcription was inaccurate.

Give the descriptive transcript its own visual wording

The original descriptive transcript draft is:

Planned source PLAN-V1. Speaker: “Choose the labeled folder.” Visual description: The sketch moves into the folder labeled Drafts. Sound: A completion chime. Speaker: “The move is complete.”

Label the visual description separately from the speaker's words. This text preserves the destination and action without pretending they were heard in the original audio.

A descriptive transcript combines the needed audio information and relevant visual description as text. It is useful to review alongside the storyboard, but its existence does not prove that an actual player delivers spoken description or meets every user's access needs.

When preparing your production script, keep each description connected to its source interval, even if the finished transcript reads as a paragraph. Actual wording should be checked against the produced media: if the folder label or action changes, the old text is no longer a current description.

Revise the relevant alternative

Select an interval to inspect its audio and visual records. Change the planned folder label and the previous alternatives become stale. Rebuild planning drafts uses that supplied label in the visual description without pretending the speaker said it. In the silent interval, the V1 caption draft still says no audio cue planned.

Choose V2 narration proposal to draft “Move the sketch into the folder labeled Drafts” as newly authored speech. Its planned audio-caption content now includes that line, but no recording, cue timing or player exists. Its final duration is unknown; the 12-second V1 storyboard cannot establish it.

These are storyboard drafts. This HTML review does not create a video, audio, WebVTT file or described player. Once media exists, compare exact sound and visual action with the alternatives, then test actual keyboard controls and delivery in the intended player.

Copy a prompt for this example

Create a single self-contained HTML media-accessibility planning inspector with embedded CSS and JavaScript. Use interval buttons, a destination-label input, separate audio and visual panels, caption and descriptive-transcript drafts, and a clear Planned status. Use the selected interval to keep each description connected to its source action.

Use PLAN-V1, a fictional 12-second sorting demonstration. From 0 to 4 seconds, the speaker says “Choose the labeled folder” while a sketch and folders are visible. From 4 to 8 seconds, the sketch moves into the folder labeled Drafts without planned speech or meaningful sound. From 8 to 12 seconds, include a completion chime and “The move is complete.”

Use interval-specific visual descriptions: sketch and labeled folders for Opening, the move and destination for Silent action, and the finished state for Completion. The V1 caption draft follows the planned speech and meaningful sound. Its silent middle interval must say no audio cue is planned. Keep “The sketch moves into the folder labeled Drafts” in a separately labeled visual description, not fabricated spoken dialogue.

Allow the folder label to change. Mark all dependent drafts stale until Rebuild planning drafts is clicked. Rebuilding updates the visual description using the new label without changing V1's silent audio content.

Provide a PLAN-V2 narration proposal adding “Move the sketch into the folder labeled [current label].” Include that proposed speech in V2's caption draft. Its final duration remains unknown until the revised recording is measured.

Reset restores Drafts, V1, and the first interval. Keep all outputs Planned and explain that no video, recording, caption file, or player has been produced. Return only the complete HTML file.

Before recording the revised script, display the HTML planning inspector in NOS. Inspect the silent interval, decide how the destination will be conveyed, and carry that decision into the recording script and review it against the produced clip.

Keep reading