Applying Automatic Translation to Scanned PDF Documents

There are times when we scan paper documents and save them as PDF files in our operations. Additionally, we may receive such PDF files from our business partners. If you want to automatically translate these types of PDF files, what methods are available?
In this article, we will introduce some reasons why PDF translation may not work well and some solutions. If you are looking to streamline your translation tasks during business operations, please consider this as a reference.

The PDF file contains scanned documents stored as images, and the text data cannot be extracted (this PDF file is referred to as an "image PDF" or "scanned PDF"). Additionally, many automatic translation services only support text translation. Therefore, traditionally, a process called OCR, which converts images to text, was necessary before automatic translation. Furthermore, the translated document files often do not maintain the original layout.
To restore the layout to match the original data, editing work is required. Because of these necessary steps, it is not possible to proceed smoothly with PDF translation.

The Google Translate app features a function called "Real-time Camera Translation." When you launch the app and point your smartphone's camera at the material you want to translate, it can automatically translate while maintaining the layout. It is very convenient, but the Google Translate app is a free service, and the confidentiality of translated documents is not guaranteed. According to the terms of use, there is a possibility that data may be reused. Therefore, there are issues with using the Google Translate app to translate business documents. Data leaks from highly confidential documents used in business settings can directly impact a company's credibility, so caution must be exercised in its usage.

How to Use the Google Cloud Translation API

This month, the document translation feature of the Google Cloud Translation API has been launched. By using this API, you can achieve functionality equivalent to the real-time camera translation of the Google Translate app, allowing for automatic translation of image PDFs while maintaining layout. Additionally, since Google does not reuse the data, the confidentiality of the data is also preserved.

I quickly tried the document translation feature. First, I printed our English homepage and scanned the paper to create an image PDF. The image below is an excerpt from that image PDF.

This was automatically translated into Japanese using the Translation API's document translation feature.

As you can see, automatic translation was achieved while maintaining the layout. The sentences are correctly recognized, and the translation accuracy is high.

How to Translate PDF Files Using DeepL

"DeepL" is a translation service provided by DeepL. It is highly regarded for its translation accuracy and supports 26 languages, including Japanese, English, and German. Compared to Google Translate, which supports over 100 languages, it may seem to have fewer supported languages, but it covers most of the languages commonly used in business settings, making DeepL sufficient for many needs. The usage is very simple. After accessing the DeepL site, select "Text Translation," paste the text you want to translate, and choose the desired target language to automatically get the translation.

DeepL translation has both free and paid versions. Since they differ in aspects such as security and the number of characters that can be translated, please choose the plan that best suits your usage when using the service.

If you wish to use the paid version, "DeepL Pro," you can either apply directly to DeepL or choose the MTrans for Office service, which is equipped with the DeepL translation engine.
The differences between DeepL's free and paid versions (DeepL Pro) are explained in detail in this article.
> What are the differences between DeepL's free and paid versions (DeepL Pro)? – Pricing, Security, Character Limits –

How to Translate PDFs with MTrans for Office

If you want to safely use the paid version "DeepL Pro" or Google's translation engine equipped with high-precision OCR functionality, one option is to use the automatic translation software "MTrans for Office."
MTrans for Office is a plugin that adds translation functions to the Word, Excel, PowerPoint, and Outlook applications you normally use.

The Windows version of MTrans for Office is equipped with a PDF translation feature that supports not only text-based PDFs but also scanned PDFs saved as images.
Since both DeepL and Google can be used during translation, you can switch between translation engines according to the document’s content and purpose, compare each translation result, and choose the most suitable one.
Additionally, the translated file is saved as an editable PDF that retains the original layout, significantly reducing the effort required for subsequent corrections and layout adjustments.

Furthermore, since MTrans for Office uses an API connection, unlike free automatic translation services, the input data is not reused for other purposes. Therefore, even for highly confidential business documents, you can proceed with translation work with peace of mind without worrying about information leaks.
MTrans for Office also offers a 14-day free trial. Please feel free to contact us.

Features of MTrans for Office

① Unlimited number of translatable files and glossaries with a flat-rate plan
② One-click translation from Office products!
③ Secure API connection
・For customers who want further enhancements, SSO and IP restrictions are also available

④ Japanese-language support by a Japanese company
・Support for security check sheets is also available
・Payment by bank transfer is available

MTrans for Office is an easy-to-use translation software for Office.