Hey Product Hunt 👋 After running Video2Any in the wild for two months, it feels like the right time to properly introduce it here. During that time, I’ve heard from many people using it to turn lecture and meeting recordings into PowerPoint decks, PDFs, images, and subtitles. The most rewarding part has been seeing it save people from scrubbing through a long recording and taking screenshots one by one. I built Video2Any because I got tired of every product needing an AI chatbot or agent. This particular job doesn’t need one. When you want the slides that actually appeared in a recording, waiting for an agent to process the video, spend a pile of tokens, and generate a different presentation is often slower—and the result may not be what you wanted. Video2Any takes a more direct approach. It is built on top of my open-source frame-difference detector, video-slide-extractor. It scans the video for meaningful visual changes, extracts the distinct slides, and gives you a review grid before export. This works particularly well for lectures, webinars, screen recordings, and slide-heavy meetings. There are no prompts and no token bill. On a typical laptop, a one-hour lecture often takes around 40 seconds to scan, although the exact time depends on the device, browser, and video format. For local files, the processing happens inside your browser, so the video stays on your device. You don’t need an account to start. If you try it, I’d especially love to know: What kind of recording did you use? What did Video2Any miss or detect incorrectly? Which export format did you actually need? I’ll be here throughout launch day, reading every comment and fixing what I can. Thanks for taking a look.

Video2Any

Extract original slides from video—without uploading it

Extract original slides from video—without uploading it