Extract text: settings and best results

What does Extract text do?

Extract text reads the words in an image and saves them as a plain text file, on your own computer. In the app the tool is called Extract Text. Point it at a screenshot, a photo of a menu or a receipt, or a scanned letter, and you get text you can paste into an email, a document or a spreadsheet.

Each image gives one text file. There is no PDF and no picture in the result, only the words. If you want a searchable PDF instead, use Image to PDF, called Image OCR in the app.

01Step 1: Open Extract Text and add your images

Click Tools in the sidebar, open the Image tab and click Extract Text in the Text section. You can also type "extract text" in the search box at the top.

Drop your images on the queue, or click it to pick files. JPG, PNG, WebP, AVIF, HEIC, HEIF, TIFF, GIF and BMP files work. Every image gets the same settings. To add more later, use Add more.

12345

1. 1Add more
2. 2Search languages
3. 3English, built in
4. 4OCR · English
5. 5Run extract

03Step 3: Run, then copy the text

The line above the button sums up the run, for example OCR · English. Check where the files will go: Default output folder saves them in your workspace's output folder, in an extract_text folder with a subfolder for today's date, such as 10-15. Click Change to pick another folder for this run. Then click Run extract.

When the run ends, each image shows Done. Click View to see the text. The name of the text file is at the top. Click Copy text to copy all of it, or select a part with your mouse and copy that. When you ran several images, use the previous and next arrows at the top to move between their text files.

123

1. 1The text file
2. 2Copy text
3. 3Previous and next file

Each text file is named after its image with -text added, such as juniper-autumn-menu-text.txt. Open folder takes you to the files. For more on the results window, see See and compare your results.

1234

1. 1Done
2. 2View
3. 3Saved to
4. 4Open folder

Every setting, and when to change it

Extract Text has one setting and no presets.

Languages

The languages the text is written in. English is built in; every other language is a one-time download.

• Range37 languages, several at once
• DefaultEnglish

How do I get the most accurate text?

1. Use a sharp image. For photos, shoot straight on, in even light, with the text filling the frame. Extract Text does not straighten a tilted page or remove shadows.
2. Crop to the part you need, for example with Crop image. Less clutter means less stray text. Sideways photos are fine: Hangar turns them upright using the orientation your camera saved.
3. Use the full-size image. Small images have small letters, and small letters are guessed.
4. Tick the right languages, and only those. Accents and special letters are read correctly only when their language is ticked.
5. Read the result in View before you paste it anywhere, especially numbers and prices.

Why is the text wrong or missing?

• The photo is blurry, dark or taken at an angle. Take it again straight on, in good light.
• The image is small. A thumbnail or a downsized photo has tiny letters. Use the original file.
• The wrong language is ticked. A French menu read as English loses its accents and some words. Tick the language and run again.
• The text is vertical. Chinese, Japanese and Korean are read when the text runs across the image, not in vertical columns.
• The file is empty or almost empty. No readable text was found. Check that the image really contains printed text, and try a sharper copy.

Troubleshooting

• Run extract is grayed out and the panel asks you to download a language. A ticked language is not downloaded yet. Click Download next to it, or untick it.
• A file failed because the image is too large. Images over 80 megapixels are not accepted. Make it smaller first with Resize images, then run it again.
• A file failed because Hangar couldn't read the image. The file may be damaged or not really an image. Open it in another app and save it again as JPG or PNG.
• A file failed while the text was being read. Try that image again on its own. If it keeps failing, save it again as a PNG and retry.
• A file failed with a message to reinstall Hangar. Part of the app that reads text is missing. Reinstall Hangar from the website.

For other errors, see A file failed: what to check.

Limits

• The result is plain text only. Fonts, colors, pictures and page layout are not kept.
• One text file per image. There is no single text file for the whole batch.
• Image files only. SVG and camera RAW files are not accepted. For scanned PDFs, use OCR PDF.
• Images over 80 megapixels are not accepted.
• An animated GIF or WebP is read as a single still frame.
• Works best on printed text. Handwriting is usually not recognized.
• Vertical Chinese, Japanese or Korean text is not supported.
• Languages other than English need a one-time download before the first run. After that, everything runs on your computer and nothing is uploaded.