How to Translate Text in a Screenshot on Mac (OCR + Translation)

The problem: text trapped in images

You screenshot an error message in Japanese from a vendor’s API docs. You grab a UI mockup with Chinese labels from a design file. A colleague pastes a screenshot of a German email thread into Slack. The text you need to understand is locked inside an image, in a language you don’t read.

The workflow is always the same: extract the text (OCR), then translate it. macOS can do both natively starting with Monterey, and there are faster third-party options for heavy use. Here’s how each method works.

Method 1: Live Text + right-click Translate (built into macOS)

Starting with macOS Monterey, Live Text recognizes text in images system-wide. Combined with the built-in Translate feature, you can go from screenshot to translation without installing anything:

1. Open your screenshot in Preview, Quick Look (select the file and press Space), or Photos.
2. Hover over the text in the image. When Live Text detects it, you’ll see the text become selectable (the cursor changes to a text selection cursor).
3. Select the text you want to translate.
4. Right-click the selection and choose Translate “[selected text]”.
5. A popover appears with the translation. Click Copy Translation to grab it.

macOS auto-detects the source language. The translation popover also lets you switch between languages if the auto-detection is wrong. You can translate into any language pair that macOS supports (the list grows with each release — macOS Sequoia supports over 20 languages).

Limitations: Live Text requires clear, high-resolution text. Handwriting, heavily stylized fonts, text on complex backgrounds, and very small text can fail. If Live Text doesn’t detect the text at all, try zooming into the image first — it sometimes helps with detection on smaller text.

Method 2: Live Text directly from the screenshot thumbnail

You don’t even need to open the screenshot in an app. When the floating thumbnail appears after taking a screenshot (⌘ + Shift + 3 or 4), you can:

1. Click the thumbnail before it disappears to open Markup.
2. In the Markup view, hover over text — Live Text works here too.
3. Select, right-click, and choose Translate.

This saves a step if you’re translating text from something you just captured. No need to navigate to the file in Finder.

Method 3: Shortcuts automation for one-step translation

For frequent screenshot translation, build a macOS Shortcut that chains OCR and translation into a single action:

1. Open the Shortcuts app.
2. Create a new Shortcut.
3. Add Get Text from Image (this runs OCR on the input).
4. Add Translate Text with Apple Translate (set your target language).
5. Add Copy to Clipboard or Show Result.
6. Set the Shortcut to accept Images as input.

Now you can right-click any screenshot in Finder, go to Quick Actions, and run your translation Shortcut. Or assign it a keyboard shortcut for even faster access. The entire flow — OCR to translation to clipboard — happens in about a second.

You can also set this up as a Quick Action in Finder, so it appears in the right-click menu for image files automatically.

Method 4: Safari and Chrome — translate screenshots of web content

If the foreign text is on a web page, don’t screenshot it at all — translate the page first:

Safari: Click the translate icon in the address bar (or go to the page title menu) and select Translate to [your language]. Safari translates the entire page in place. Then screenshot the translated version.

Chrome: Right-click anywhere on the page and select Translate to [your language]. Chrome uses Google Translate to render the page in your language. Screenshot after translation.

This gives you cleaner results than OCR-based translation because you’re working with the actual text, not an image of it. The translation is also more accurate because the engine has the full page context, not just an isolated snippet.

Method 5: AI assistants — screenshot + translate in one prompt

Modern AI assistants with vision capabilities (Claude, ChatGPT, Google Gemini) can read text from screenshots and translate it in a single step:

1. Take a screenshot of the foreign-language text.
2. Paste or upload the screenshot to your AI assistant.
3. Ask: “Translate the text in this screenshot to English.”

This is the most flexible method because AI handles messy input that breaks traditional OCR: handwritten notes, text overlaid on images, mixed languages in a single screenshot, unusual fonts, and low-resolution captures. The AI also provides context that a translation engine can’t — it can explain idioms, flag ambiguous translations, or note that a term is technical jargon.

The downside is speed. Pasting into a chat and waiting for a response is slower than the native right-click > Translate flow. Use AI for complex or ambiguous translations, and the native macOS method for quick, straightforward ones.

Method 6: Third-party tools for power users

If you translate screenshots regularly, dedicated tools streamline the process:

TextSniper. A Mac app that captures text from anywhere on screen with a keyboard shortcut (similar to the macOS screenshot selection tool, but for text). Select a region, and it copies the recognized text to your clipboard. Pair it with a translation service for a two-step workflow: capture text, then paste into your translator of choice. TextSniper handles stylized and angled text better than Live Text.

DeepL. The DeepL desktop app for Mac lets you paste text and get high-quality translations, especially for European languages. Combine TextSniper (OCR) + DeepL (translation) for a fast two-app workflow. DeepL’s translations are often more natural-sounding than Apple’s built-in Translate, particularly for longer passages.

Google Lens (via Chrome). Open the screenshot in Chrome, right-click the image, and select Search image with Google Lens. Lens detects text in the image and offers a translate option. This works well for screenshots with mixed text and images, like product packaging or signage photos.

Quick comparison

Tips for better OCR accuracy on screenshots

Capture at native resolution. Don’t resize or compress the screenshot before running OCR. Retina Mac screenshots are 2x resolution, which is ideal for text recognition. Downscaling before OCR reduces accuracy significantly.

Capture just the text region. Use ⌘ + Shift + 4 to select only the area containing the text you want to translate. Less visual noise means better OCR results. Full-screen captures with sidebars, toolbars, and other UI elements can confuse the text detection.

Increase contrast when possible. If you’re screenshotting a web page with low-contrast text (light gray on white), try toggling dark mode or using Reader View in Safari before capturing. High contrast between text and background dramatically improves OCR accuracy.

Check the source language. Live Text and most OCR tools auto-detect the language, but they can misidentify similar scripts (e.g., confusing Japanese and Chinese characters). If the translation looks wrong, manually set the source language in the translation step.

Install language packs. macOS Translate works offline for downloaded languages. Go to System Settings > General > Language & Region > Translation Languages and download the languages you translate most frequently. This makes translations faster and works without an internet connection.

Real-world use cases

Developers reading foreign API documentation. Many API docs, error messages, and code comments are in the maintainer’s native language. Screenshot the relevant section, translate, and move on — faster than finding an English version that may not exist.

Designers working with international content. When reviewing designs for localized versions of an app, you often need to understand what the translated UI text says. Screenshot the mockup, translate the labels, and verify that the content fits the design.

Researchers and students. Academic papers, source documents, and archived materials are often only available in their original language. Screenshot, translate, and annotate — a faster workflow than copying text manually from PDFs with broken encoding.

Customer support. When users send screenshots of error messages or UI issues in their language, support teams need to understand the content to help. The right-click Translate workflow handles this in seconds.