Image to text (OCR)
How to extract text from an image
- 1. Add an image
Drop a JPEG, PNG, WebP or BMP up to 30 MB on the box, click it to select one, or paste a screenshot with Ctrl+V.
- 2. Select the language
Auto-detect reads English and Vietnamese. For another language, select it in Image Language, and the tool reads the image again.
- 3. Wait for the text
The text shows in the box. The first run takes longer, because your browser downloads the OCR engine and the language data.
- 4. Copy or download
Correct the text in the box if necessary, then click Copy text, or Download to save a .txt file.
Frequently asked questions
Is my image uploaded?
No. The text is read by Tesseract, an OCR engine that runs in your browser. The image stays on your device. The first time, your browser downloads the engine and the language data.
Is it free? Is there a limit?
It is free, with no account and no daily limit, because your own device does the work. Each image can be up to 30 MB, and you read one image at a time.
Which languages can it read?
English, Vietnamese, Chinese (Simplified and Traditional), Japanese, Korean, French, German, Spanish and Russian. Auto-detect reads only English and Vietnamese. For another language, select it in Image Language.
Can it read Vietnamese text with diacritics?
Yes. Auto-detect and Vietnamese use the Vietnamese language data of Tesseract, which reads letters with diacritics and tone marks. Check the result, because small or blurry marks can be read wrong.
What kind of image gives the best result?
A clear, straight image of printed text, such as a screenshot or a flat scan. Handwriting, very small text, dark photos and tilted pages give more errors. You can correct the text in the box before you copy it.
Can I use a PDF?
Not in this tool, which reads image files only. For a PDF, use the Universal file converter and select Plain Text: it reads the text of the PDF, and uses the same OCR for English and Vietnamese on scanned pages.