# SlideSieve

> SlideSieve is an AI-powered browser extension for Google Chrome and Microsoft Edge that automatically captures presentation slides and extracts synced transcripts from online course videos on DeepLearning.AI, Coursera, Udemy, and YouTube. Captured slides can be exported as multi-page PDFs or individual images, and transcripts can be exported as PDF, TXT, SRT, or VTT. All processing happens 100% locally in the user's browser — no video data, slides, or personal information are ever uploaded to any server.

## What is SlideSieve?

SlideSieve watches online video lectures as you watch them and automatically detects whenever the slide on screen changes. Using structural image analysis and perceptual hashing, it captures each unique slide frame, filters out instructor talking-head segments and near-duplicate frames, and stores them neatly organized by course and lesson. In parallel, it can extract the complete lesson transcript, synchronized in real time with the video. When finished, learners can export captured slides as a formatted multi-page PDF or individual images, and export transcripts as TXT, SRT, or VTT files.

## Supported Platforms

- **DeepLearning.AI** ([SlideSieve for DeepLearning.AI](https://slidecapture.com/for-deeplearning-ai)) — Specially optimized for learn.deeplearning.ai; captures lecture slides automatically where no official slide downloads exist.
- **Coursera** ([SlideSieve for Coursera](https://slidecapture.com/for-coursera)) — Automatically detects and captures slides and synchronized transcripts from Coursera video lectures.
- **Udemy** ([SlideSieve for Udemy](https://slidecapture.com/for-udemy)) — Works seamlessly across all Udemy course players, including DRM-protected video streams.
- **YouTube** ([SlideSieve for YouTube](https://slidecapture.com/for-youtube)) — Captures slides and extracts transcripts from any educational YouTube video, lecture, or conference presentation.

## Key Features

- **Auto Slide Detection**: Structural image analysis detects genuine slide transitions without relying on video metadata, identifying presentation slides vs. talking-head segments across any course layout.
- **Instructor Webcam Filter**: Automatically identifies and skips frames where an instructor's webcam or talking head dominates the screen, ensuring only clean slide content is captured.
- **Smart Deduplication**: Eliminates near-identical frames so learners never get redundant copies of static slides, while reliably capturing progressive bullet points and step-by-step builds.
- **Slide Detection Zone**: Users can drag and draw a crop rectangle over the slide area to constrain detection or exclude floating webcam overlays. The zone persists across all lessons in the course.
- **PiP Magic Eraser**: When a zone excludes a webcam overlay, SlideSieve reconstructs the real slide background behind the camera window from a rolling frame buffer rather than merely blurring or cropping. Exported slides show zero webcam traces.
- **Transcript Extraction**: Extracts full lesson transcripts synced live to video playback, with the currently spoken sentence highlighted in real time. Features both Timeline View (timestamped) and Article View (reading paragraphs), with export to PDF, TXT, SRT, VTT, or one-click clipboard copy.
- **Manual Snap**: One-click instant capture for diagrams, terminal outputs, or code snippets at any moment, regardless of auto-detection status.
- **In-Page Floating Toolbar**: An unobtrusive floating toolbar appears directly below supported video players to toggle capture, take manual snaps, or extract transcripts without opening the extension side panel.
- **Multi-View Slide Browser**: Review captured slides in grid, list, or full-screen presentation mode, and delete unwanted frames before exporting.
- **PDF & Image Export**: Compiles captured slides into a single multi-page PDF preserving each slide's native aspect ratio, or exports individual high-resolution image files.
- **Course-Scoped Storage**: Slides and transcripts are automatically organized by course and lesson name, persisted locally in IndexedDB across browser sessions.
- **15 Languages**: The extension interface is fully localized in English, German, French, Spanish (Spain and Latin America), Italian, Portuguese (Portugal and Brazil), Japanese, Korean, Simplified Chinese, Traditional Chinese, Indonesian, Hebrew, and Turkish. Auto-detected on install and changeable anytime in Settings.
- **100% Private & Local**: All slide detection, OCR, webcam filtering, and transcript extraction execute entirely within the local browser. No personal data, course videos, or slide images are ever transmitted to external servers.

## Pricing

- **Free Plan**: $0 forever. Includes full auto slide detection, deduplication, instructor filtering, transcript extraction, and custom capture/exclude zones, with a lifetime budget of up to 150 slide captures. No credit card required.
- **Pro — 3-Month Access**: $6.99, one-time payment for 90 days of unlimited slide captures and unlimited PDF exports from purchase. Ideal for students and intensive course sprints. No recurring subscription.
- **Pro — Lifetime**: $16.99, one-time payment for permanent access. Includes unlimited slides per lesson, unlimited PDF exports, multi-format export, priority support, and all future Pro updates. No recurring subscription.

## How to Use SlideSieve

1. Install SlideSieve from the Chrome Web Store or Microsoft Edge Add-ons.
2. Navigate to a video lecture on DeepLearning.AI, Coursera, Udemy, or YouTube.
3. (Optional) Click the Region Selector on the floating toolbar to draw a Slide Detection Zone around the slide area or exclude the instructor's webcam.
4. Click "Start Auto Capture" on the floating toolbar below the video player.
5. Play the video — slide changes are detected and captured automatically.
6. Use the "Snap" button (or shortcut `Ctrl+Shift+9` / `Cmd+Shift+9`) anytime to manually capture a specific frame.
7. Open the SlideSieve side panel to review slides in grid, list, or presentation view, and delete unwanted frames.
8. Click "Export" to generate a multi-page PDF or individual images, and optionally export transcripts as PDF, TXT, SRT, or VTT.

## Official Links

- **Website**: https://slidecapture.com
- **Chrome Web Store**: https://chromewebstore.google.com/detail/slidesieve-auto-slide-ext/lbiholhfnfedgecgkfclmhnphapaopep
- **Microsoft Edge Add-ons**: https://microsoftedge.microsoft.com/addons/detail/dmjaijcodfgojhepdllkodnpkcaejoac
- **DeepLearning.AI Guide**: https://slidecapture.com/for-deeplearning-ai
- **Coursera Guide**: https://slidecapture.com/for-coursera
- **Udemy Guide**: https://slidecapture.com/for-udemy
- **YouTube Guide**: https://slidecapture.com/for-youtube
- **Pricing**: https://slidecapture.com/#pricing
- **Changelog**: https://slidecapture.com/changelog
- **Blog & Tutorials**: https://slidecapture.com/blog
- **Support**: https://slidecapture.com/support
- **Privacy Policy**: https://slidecapture.com/privacy
- **Terms of Service**: https://slidecapture.com/terms

## Frequently Asked Questions

**Does SlideSieve work on Coursera?**
Yes. SlideSieve automatically detects and captures presentation slides and extracts synchronized transcripts from Coursera video lectures, organizing them by course and lesson name into downloadable multi-page PDFs.

**Does SlideSieve work on Udemy?**
Yes. SlideSieve is fully compatible with Udemy's video player, including DRM-protected video content. It captures clean slide frames automatically while ignoring instructor webcam overlays.

**Does SlideSieve work on YouTube?**
Yes. SlideSieve operates on any educational YouTube video, webinar, or recorded conference lecture, automatically capturing slide changes and extracting synced subtitles and transcripts.

**Does SlideSieve work on DeepLearning.AI?**
Yes. SlideSieve is specially optimized for the learn.deeplearning.ai platform. Because DeepLearning.AI does not provide official slide downloads for its Short Courses, SlideSieve captures these slides automatically as you watch, eliminating the need for manual screenshots.

**Which browsers does SlideSieve support?**
SlideSieve is officially available for Google Chrome (via the Chrome Web Store) and Microsoft Edge (via Microsoft Edge Add-ons). It also runs smoothly on other Chromium-based browsers including Brave, Opera, and Arc.

**Can SlideSieve extract video transcripts and subtitles?**
Yes. SlideSieve extracts full lesson transcripts across all supported platforms, synchronized live with playback and highlighting current sentences. Transcripts can be reviewed in Timeline View or Article View and exported as PDF, TXT, SRT, or VTT files, or copied directly to clipboard.

**Can SlideSieve remove the instructor's webcam from captured slides?**
Yes. When using a Slide Detection Zone to exclude a Picture-in-Picture webcam overlay, SlideSieve's PiP Magic Eraser reconstructs the genuine slide background behind the webcam by sampling clean background pixels from a rolling frame buffer, leaving no visible watermark or blur.

**Can I capture slides manually?**
Yes. You can click the "Snap" button on the in-page floating toolbar or press `Ctrl+Shift+9` (`Cmd+Shift+9` on Mac) at any time to instantly save the current frame, regardless of whether auto-detection is active.

**What keyboard shortcuts does SlideSieve support?**
- `Ctrl+Shift+8` (or `Cmd+Shift+8` on Mac): Quickly toggle Auto-Capture on or off.
- `Ctrl+Shift+9` (or `Cmd+Shift+9` on Mac): Manually snapshot the current video frame instantly.
Shortcuts can be customized anytime via `chrome://extensions/shortcuts` or `edge://extensions/shortcuts`.

**Can I switch tabs while Auto-Capture is running?**
Auto-Capture pauses automatically when switching to another tab or minimizing the browser, because modern browsers throttle background video rendering. Keep the course tab visible for continuous recording. Capture resumes instantly once you switch back without losing progress.

**Is SlideSieve free?**
Yes. SlideSieve offers a free tier with no credit card required, including auto slide detection, transcript extraction, and a lifetime quota of up to 150 slide captures. Paid options include 3-Month Access ($6.99, one-time payment for 90 days from purchase) and Pro Lifetime ($16.99, one-time payment for permanent access), both providing unlimited captures and exports with no recurring subscriptions.

**Is my data private with SlideSieve?**
Yes, 100%. All slide analysis, image processing, and transcript handling occur strictly on-device in your browser. No video streams, slide captures, transcripts, or personal analytics are ever transmitted to external servers.

**What languages is SlideSieve available in?**
The extension UI is localized into 15 languages: English, German, French, Spanish (Spain and Latin America), Italian, Portuguese (Portugal and Brazil), Japanese, Korean, Simplified Chinese, Traditional Chinese, Indonesian, Hebrew, and Turkish. The language is auto-detected on install and can be adjusted anytime in Settings.

## Trademark Notice

SlideSieve is an independent software product and is not affiliated, associated, authorized, endorsed by, or in any way officially connected with Coursera, Udemy, YouTube, DeepLearning.AI, or Google LLC. All product names, logos, and trademarks mentioned are the property of their respective owners and are used strictly for identification and descriptive educational purposes under fair use.

## Version History

- **v1.1.0** (August 2026): Major release introducing the dedicated Transcript Tab with automatic transcript extraction across all supported platforms; live-synced Timeline View and Article View; TXT, SRT, VTT, and clipboard export; Slide Detection Zone with drag-to-draw overlay and PiP Magic Eraser background reconstruction; 15-language UI support; redesigned export modals; and improved duplicate detection stability.
- **v1.0.9** (August 2026): Full UI internationalization with custom settings language selector, automatic locale detection on first run, refined tooltip typography, and fix for tooltip clipping on long translated text.
- **v1.0.8** (August 2026): Complete dark mode UI/UX overhaul with WCAG AA compliant contrast ratios, elevated control surfaces, refined license input styling, and enhanced Unlock Pro cards.
- **v1.0.7** (August 2026): Added custom course selection dropdown with platform badges, updated official extension title to emphasize auto-capture, and removed external CDN references for strict Manifest V3 compliance.
- **v1.0.6** (August 2026): Switched to a fairer 150-slide lifetime free budget with persistent sync storage, real-time usage progress indicator, and instant capture shutoff upon reaching the limit.
- **v1.0.5** (July 2026): Chrome Web Store international SEO expansion across 65 locales, added safe-harbor trademark disclaimer notices, and introduced interactive 5-star rating component in Settings.
- **v1.0.3** (June 2026): Real-time capture interval timer adjustments that apply dynamically to ongoing recording sessions.
- **v1.0.2** (June 2026): Multi-region slide detection supporting diverse slide aspect ratios and logo placements; fixed iframe player traversal and DeepLearning.AI edge-case capture failures.
- **v1.0.1** (June 2026): Dynamic course and lesson title extraction for automated PDF file naming, individual slide aspect ratio preservation in PDF export, instructor talking-head human veto, and letterbox/pillarbox bounds detection.
- **v1.0.0** (June 2026): Initial public release of SlideSieve with automatic slide detection, deduplication, and PDF export for online video lectures.