When you need the text from a screenshot, scan, or photo as copyable text, this OCR (optical character recognition) tool runs entirely on your device. Drop an image in (or click to choose, or paste) and press Extract text — an in-browser AI OCR model (PP-OCRv5) finds the text lines and recognizes them, showing the result in the text box, ready to copy for translation, search, or editing. There is no language picker: Japanese, English, and Chinese — even mixed in one image — are recognized automatically. Importantly, your image is never uploaded, stored, or sent anywhere: recognition runs entirely inside your browser, so you can transcribe internal documents, paperwork, or images with personal data safely. The model loads only on first use (about 11 MB) and runs quickly from cache afterwards; on supported browsers it uses the GPU (WebGPU). For best accuracy, use a high-resolution image that is sharp and high-contrast — slight tilt is corrected automatically. It is built for printed/on-screen text; cursive handwriting and highly decorative fonts are harder. For PDFs that already have a text layer, the 'PDF Extract Text' tool is more accurate (OCR works on a rasterized image of the PDF).