Recap Audio Describer

Recap Audio Describer generates automated extended audio descriptions for web video, entirely in your browser. No servers, no uploads, no data leaves your device. Built for accessibility teams, instructional designers, government agencies, and anyone who needs to make video content accessible to people who are blind or have low vision. Please note that this extension is currently in a beta/early proof-of-concept stage; the enterprise application at voicerecap.ai is much more battle-tested and proven. HOW IT WORKS 1. Navigate to any web page with a video. 2. Click the Recap Audio Describer icon in your browser extension toolbar. 3. The extension scans the page and lists every video it finds. Click "Describe" next to the one you want. 4. The video plays while AI analyzes each scene in the background. Descriptions appear in a side panel as they are generated, each with a thumbnail and timestamp. 5. Descriptions are generated in three verbosity formats; short, medium, and long. 5. When all descriptions are ready, click "Play all" to replay the video with spoken descriptions inserted at each cue point. The video pauses briefly while each description is read aloud, then resumes automatically. 6. Export your completed descriptions as SRT, WebVTT, plain text, or print a formatted transcript with thumbnails and timestamps. COMPLETELY PRIVATE, ON-DEVICE AI All AI processing happens locally on your computer using your browser's built-in AI model. Video frames are captured and analyzed without ever leaving your device. No cloud APIs, no third-party AI services, no video data transmitted anywhere. Your content stays yours. Subscription billing processed via Stripe. If audio description tracks are exported (not required), in order to facilitate notification when finalized, described track data may be sent through our email infrastructure. EXTENDED AUDIO DESCRIPTIONS (AD) Unlike inline standard audio descriptions that must fit between dialogue, Recap Audio Describer produces extended descriptions. The video pauses at each cue point to allow the full description to be read before resuming playback. This provides richer, more detailed scene information for viewers who are blind or have low vision, following the extended audio description methodology recommended by accessibility standards. ACCESSIBILITY COMPLIANCE Audio descriptions are a legal requirement under multiple accessibility frameworks. Recap Audio Describer helps your organization meet compliance for video content across: - Americans with Disabilities Act (ADA) - Section 508 of the Rehabilitation Act - Web Content Accessibility Guidelines (WCAG) 2.1 AA (Success Criterion 1.2.5 and 1.2.7) - Accessibility for Ontarians with Disabilities Act (AODA) - European Accessibility Act (EAA / EN 301 549) - Communications and Video Accessibility Act (CVAA) Whether you are a state agency publishing public meeting recordings, a city council with planning commission meetings, a university posting lecture videos, or a media company distributing content online, audio descriptions are increasingly expected or required. Recap Audio Describer lets you generate them automatically. VERBOSITY CONTROL Choose from three description lengths to match your content and audience: - Short: One concise sentence per scene. Best for fast-paced content or quick drafts. - Medium: One to two sentences with additional context. The default setting and a good balance for most videos. - Long: Two to three detailed sentences describing visual elements, actions, and setting. Best for instructional or narrative content where detail matters. All three verbosity levels are generated simultaneously for each scene, so you can switch between them instantly without re-processing. VOICE SELECTION Pick the voice that works best for your audience. The extension lists every text-to-speech voice available in your browser, with standard and basic voice groups clearly labeled. Preview any voice before committing. The selected voice is saved and remembered across sessions. A wider range of premium voices and dialects that are much richer, and more natural/human-sounding are available to our enterprise clients through voicerecap.ai. EXPORT AND SHARING Once descriptions are generated, export them in the format your workflow requires: - SRT: Industry-standard subtitle format compatible with most video players, editors, and media platforms. - WebVTT: Web-native subtitle format with metadata indicating AI generation date and source. Ready for use with HTML5 video players. - Plain text (TXT): Timestamped transcript with generation details. Useful for documentation, review, or archival. - Print: Formatted printable document with video thumbnails at each cue point, timestamps, and Recap Innovations branding. Ideal for review packets, compliance documentation, or stakeholder sign-off. Export filenames automatically include the video title and a clear identifier so you can organize your described content. KEYBOARD SHORTCUT Press Alt+D on any page with a video to instantly start describing. The side panel opens alongside the page so you can continue browsing while descriptions are generated. DESCRIPTION MANAGEMENT Every generated description appears in a scrollable history panel with: - A thumbnail snapshot of the video at that moment - The timestamp in the video - The full description text - A button to replay the spoken description - A button to seek the video to that point - Options to regenerate or delete individual descriptions Descriptions persist across verbosity changes. If you switch from medium to short and back, your original descriptions are still there. SYNC PLAYBACK Click "Play all" to replay the video synchronized with your generated descriptions. The video plays naturally between cue points, pauses while each description is spoken aloud with audio ducking (the video volume lowers so the description is clearly audible), then resumes automatically. A completion chime sou…