About VidText

What is VidText?

VidText is a Chrome browser extension that lets you copy text from any video on any webpage. Simply pause a video, and VidText uses on-device AI (PP-OCRv5 via WebAssembly) to instantly extract and overlay selectable text — pixel-aligned to its original position on the video frame.

Unlike cloud-based OCR tools, VidText processes everything locally in your browser. Your video frames never leave your device. No accounts, no API keys, no data collection — just a simple tool that works.

Who is it for?

VidText is built for anyone who needs to extract text from video content:

• Students — Copy lecture slides, code snippets, and notes from online courses.
• Developers — Grab code examples and terminal output from tutorial videos.
• Researchers — Extract text from video presentations, documentaries, and archival footage.
• Knowledge workers — Quickly capture information from webinars, meetings, and training videos.

Technology

VidText is powered by PP-OCRv5, an advanced OCR model from PaddlePaddle, running entirely in your browser via ONNX Runtime Web and WebAssembly with SIMD acceleration. The extension uses Chrome's Manifest V3 architecture, Shadow DOM for UI isolation, and multi-threaded inference for optimal performance.

Supported Platforms

VidText works on any webpage with HTML5 video elements, including YouTube, Vimeo, Udemy, Coursera, Kaltura, SharePoint, and many more. It also supports videos embedded in iframes.

Contact & Support

Have a question, suggestion, or bug report? We'd love to hear from you.