Upload screenshots to help translators with context and harness the magic of OCR to automatically link your screenshots to the related content. In this article, we’ll explain how we’ve made localization and translation easier at Localazy by incorporating the OCR technology into our Context Screenshots feature.
🧐 What is OCR?
Optical Character Recognition (OCR) is a technology that interprets images of handwritten or printed text and converts them into a format suitable for input to computers. This technology is used mainly in libraries, large corporations, and other places that need to convert printed text into machine-encoded text regardless of its source.
The technology finds use in data entry tasks, allows time-effective extraction, and assists in ranking and indexing web pages with printed materials on search engines. However, it can also be used in translation management systems like Localazy to smoothen the localization process.
OCR technology is not limited to English but can detect any text. It can be employed to extract information from different files as long as the fonts used are comprehensible. 🔍
📸 Introducing Context Screenshots
Context runs the show in the translation scene. A sure way to boost the accuracy of the translations is to find ways to simplify how context can be provided. Localazy allows project maintainers using the Autopilot plan to add screenshots and link them with the respective source texts from your project for members and contributors to refer to. 🔗
Since most strings are made of single words, it’s difficult to translate them to fit the context when there’s no visual representation. By utilizing Screenshots, translators aren’t left to their imagination to decide how the interface might look and to which part of the project the translated content belongs. With screenshots, translators know exactly where the strings fit into the software or app’s interface.
If the project you’re working on has many individual interfaces, then it will require taking lots of screenshots which could get overwhelming if you don’t organize them. You can do this easily by tagging them with specific words and grouping them. 🏷️
👁️🗨️ Reviewers, translators, and managers can access these screenshots to ensure the translations are appropriate. But do you have to link source keys after manually checking the screenshots? Wait, did you forget that we call ourselves Localazy?
The text extraction technology mentioned above (OCR) scans the image for possible phrases that match the source keys in your project. Snap! Just like that. No manual typing is to be done from your side. And most of them are first shot, right into the spot! 🎯
Source: Giphy⏭️ Then, all you have to do when uploading these screenshots is to select whether the suggested source keys are accurate. Surely, if you’d like to soar towards the sky with your project with OCR Screenshots, you’d need the Autopilot plan to access it.️
What are the Benefits of Using Screenshots? 🤔
Pictures are worth a thousand words. Instead of overwhelming members with comments on individual strings, attach a screenshot.
Improved Context for Translators
Translators won’t be blindly translating content that they have no idea where fits on the app or website. Screenshots will give them a thorough grasp of the interface and the exact functionality of certain phrases, buttons, placeholders, etc. Through screenshots by developers or users, translators can see interface states they can’t always access.
Modal Windows and Error Messages
If a translator is instructed to use an app or software, he doesn’t have access to all its interfaces or dialogue boxes. For instance, the software might show different error messages and modal windows depending on the user, device, location, etc. 📍
🧑💻 When software makers who can trigger all these states attach screenshots, it’s easier for translators and reviewers to produce quality translations.
Context for the User-dependent States
🔑 Again, even if you’re given access to the software as a translator, the interface you see changes based on your role as a user. For instance, if you’re localizing a vacation rental software like Airbnb, logging in as a host shows other interface components compared to a standard user interface. 👀🤷🏻♂️
👍 How to Enable Screenshots OCR?
To enable the Screenshots feature, activate the best-selling Autopilot plan. Screenshots are also included in the higher tiers.
Uploading screenshots
To upload screenshots, you have two options.
- Upload manually (learn how to do it in the docs)
- Import screenshots from Figma (available in the Agency plan)
Learn How to connect Figma designs and translations via screenshots.
Linking screenshots to source keys
As mentioned above, our OCR can recognize text in the uploaded images and suggest the relevant source keys to link if there is a good match.
🕵️♂️ Localazy OCR suggests appropriate source keys that can be linked to the screenshot, and then it’s up to you to either accept the suggestions or reject them. Visit the screenshot’s detail page and tick the source keys that match your screenshots. ☑️
🤨 How accurate is the OCR feature for linking to the right source keys? We let you judge. Sometimes the OCR might miss, depending on the resolution and other aspects of the uploaded screenshot, such as contrast, visual effects affecting the texts, etc. But there is no doubt that looking up each presented key in the screenshot would be very time-consuming, so any help is welcome. And in the end, the decision of which keys belong to which screenshots are yours.
✔️ Conclusion
Even if the entire translation and localization process may seem daunting at first, never forget one crucial thing: translators will always need the best context to minimize the risk of the meaning being lost in translation.
Would you like to try the Screenshots feature and see it in action? Activate the Autopilot plan now and see for yourself!
This is a companion discussion topic for the original entry at https://localazy.com/blog/how-to-use-ocr-to-achieve-high-quality-translations