Help us improve
Share bugs, ideas, or general feedback.
From grimoire
Generates captions, transcripts, and audio descriptions for video, audio, and time-based media following WCAG 2.1 Level A/AA accessibility standards.
npx claudepluginhub jeffreytse/grimoire --plugin grimoireHow this skill is triggered — by the user, by Claude, or both
Slash command
/grimoire:apply-media-captionsThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Provide synchronized captions for video, transcripts for audio, and audio descriptions for visual-only content.
Designs accessible video/audio with captions, transcripts, and audio descriptions. Guides creation and review of multimedia for accessibility compliance.
Generates WCAG 2.1 / Section 508 compliant corrected captions for YouTube videos using AI, with configurable speech-to-text source.
Generate professional voiceover narration for a video with audio-video sync using Azure TTS by default, or Gemini 3.1 Flash TTS when configured. Use this skill whenever the user wants to add narration, voiceover, commentary, or voice dubbing to any video file — even if they just say "add audio to this video" or "make a narrated version." Also trigger when the user has a screen recording, demo, tutorial, or presentation video that needs a voice track. Trigger on Chinese requests like "视频配音", "给视频加旁白", "录屏解说", "视频加语音", "视频添加声音", "生成视频旁白", "自动配音", "视频解说词".
Share bugs, ideas, or general feedback.
Provide synchronized captions for video, transcripts for audio, and audio descriptions for visual-only content.
Adopted by: WCAG 2.1 SCs 1.2.1–1.2.5 are Level A/AA — legally required by Section 508 (US), ADA, EU EN 301 549, and the UK Equality Act. Netflix was sued and settled for $755,000 in 2012 for missing captions (NAD v. Netflix) — the landmark case that accelerated industry adoption. All major streaming platforms (YouTube, Vimeo, AWS Elemental MediaConvert) now provide auto-captioning as baseline infrastructure. Impact: WHO estimates 1.5 billion people have hearing loss. Captions also benefit users in noisy environments, non-native language speakers, and users with cognitive disabilities. Facebook data showed 85% of videos are watched without sound — captions improve engagement for all users, not just those with disabilities. Why best: Auto-generated captions alone are insufficient — accuracy rates of 80–90% produce too many errors for compliance. Human-verified captions or caption editing is required for WCAG conformance.
Sources: W3C WCAG 2.1 SC 1.2.1–1.2.5 (2018); NAD v. Netflix settlement (2012); Facebook Captions research (2016); W3C WebVTT specification
| Media type | Level A requirement | Level AA addition |
|---|---|---|
| Pre-recorded audio-only | Transcript | — |
| Pre-recorded video-only | Audio description OR transcript | — |
| Pre-recorded video+audio | Captions | Audio description |
| Live video+audio | Live captions | — |
<video controls>
<source src="product-demo.mp4" type="video/mp4">
<track kind="captions"
src="product-demo.en.vtt"
srclang="en"
label="English"
default>
</video>
WebVTT format:
WEBVTT
00:00:01.000 --> 00:00:04.000
Welcome to the product demo.
00:00:04.500 --> 00:00:08.000
Today we'll cover the three main features.
NOTE Speaker change
00:00:08.500 --> 00:00:12.000
[Jane] First, let's look at the dashboard.
Caption quality requirements:
[Speaker Name][applause], [alarm beeping]<audio controls src="podcast-ep-42.mp3"></audio>
<details>
<summary>Transcript — Episode 42</summary>
<p><strong>Host:</strong> Welcome to episode 42...</p>
<p><strong>Guest:</strong> Thank you for having me...</p>
</details>
Transcripts must include all speech, speaker identification, and meaningful non-speech sounds. A transcript linked adjacent to the player satisfies WCAG 1.2.1.
When video conveys information through visuals not described in dialogue, add an audio description track:
<video controls>
<source src="tutorial.mp4" type="video/mp4">
<track kind="captions" src="tutorial.en.vtt" srclang="en" label="English captions" default>
<track kind="descriptions" src="tutorial.en.desc.vtt" srclang="en" label="Audio descriptions">
</video>
Or provide a separate audio-described version as a link adjacent to the player.
Checklist:
aria-hidden="true" and muted autoplay. No captions required.Shipping auto-generated captions without review. Auto-caption accuracy of 80% means 1 in 5 words is wrong. This fails WCAG 1.2.2 and produces an unreliable experience.
Captions that only cover speech. A video where an alarm sounds and the character reacts — but the captions say nothing about the alarm — fails to convey equivalent information.
Transcript behind a paywall or login. If the media itself is accessible without login, so must the transcript be.