How to Translate With DeepL

It can be difficult for authors, educators, and other professionals to translate content between languages while maintaining its quality, tone, and meaning. Manual translation is expensive and time-consuming, whereas the free online tools produce vague phrasing that misses the context.

In this article, we will explore what DeepL is, walk through a complete translation workflow step by step, and learn how to use its tools to improve both translations and writing.

What is DeepL?

DeepL is a neural machine translation and AI writing tool launched in 2017 that uses advanced transformer architecture to produce context-aware translations with natural phrasing and a polished tone. It interprets entire sentences and paragraphs to comprehend context, idioms, and intended meaning, in contrast to traditional translation tools that operate word-by-word. This produces translations that sound less robotic and more human. The platform presently supports over 30 languages and is regarded for its capacity to preserve tone and pick up on linguistic subtleties.

With its core tools, DeepL accomplishes two main goals:

• DeepL Translator handles text and file translation for Word, PowerPoint, and PDF documents while preserving formatting, ideal for academic work, business documents, and marketing materials.

• DeepL Write refines English text by adjusting tone between formal and simple styles, perfect for professional emails, reports, and content creation.

Now that we know what DeepL offers, let’s understand how to use it.

How to use DeepL?

You can access DeepL in two ways:

• Visit the DeepL website through any web browser to use the online version instantly without installing anything.

• Alternatively, download the DeepL desktop app for Windows or Mac from the official website for offline access and better integration with your workflow.

In this tutorial, we are using the downloaded version of DeepL.

With the installation done, let’s implement the complete translation workflow step by step.

Implementing the translation workflow in DeepL

Let’s use DeepL to achieve a real-world goal. Consider yourself a marketing manager introducing a productivity app in France and Germany with an announcement email, PDF feature guide and brand language that needs to be maintained.

We’ll translate the email from English to French and German, process the PDF, and create a glossary to maintain your brand voice across both languages.

Step 1: Translate text from one language to another

Open the DeepL app or website and paste the following product announcement email into the source language box:

Subject: Introducing TaskFlow – Your New Productivity Partner 

Dear Valued Customer,

We're excited to announce the launch of TaskFlow, our innovative productivity app designed to streamline your daily workflow. TaskFlow helps teams collaborate seamlessly with real-time updates, smart task prioritization, and intuitive project tracking. 

Starting today, TaskFlow is available for download on iOS and Android. Try our free 30-day trial and discover how TaskFlow can transform the way your team works together. 

Best regards, 
The TaskFlow Team 

Copy to clipboard
Copy to clipboard

Select French as your target language. Here is a sample that DeepL generated instantly:

Repeat this process with German as the target language to create your second translation.

We can also choose the formality for the translated text to switch between formal and informal tones. For business communication, the formal tone is typically more appropriate in both French (vous) and German (Sie).

Step 2: Translate the documents

We have this TaskFlow Feature Guide PDF that we’ll translate to French and German. This document includes TaskFlow’s introduction, core features like real-time collaboration, smart task prioritization, advanced capabilities, and pricing plans.

To translate this PDF, go to the Translate Files tab and upload the TaskFlow feature guide.

Select French as your target language and select “Translate into French”.

DeepL will take some time to process the uploaded file and then translate it. Once done, it will generate a downloadable file that is the translated version of the original file.

Repeat this process with German as your target language.

Your translations are ready, but what if certain terms need to remain consistent across all languages? That’s where glossaries come in.

Step 3: Use a glossary for consistency

A glossary ensures that specific terms such as brand names, product features, or technical terminology are translated exactly the way you want them across all languages. For our TaskFlow launch, we need to keep “TaskFlow” as is, translate “productivity app” consistently, and maintain specific feature names.

In DeepL, select Options, then click on Glossary, and choose “Create Glossary”. Name it “TaskFlow Brand Terms” and add the following key terms with your preferred translations:

For English to French add pairs:

• TaskFlow and TaskFlow (unchanged)

• productivity app and application de productivité

• real-time collaboration and collaboration en temps réel

• smart task prioritization and priorisation intelligente des tâches

For English to German add pairs:

• TaskFlow and TaskFlow (unchanged)

• productivity app and Produktivitäts-App

• real-time collaboration and Echtzeit-Zusammenarbeit

• smart task prioritization and intelligente Aufgabenpriorisierung

Save your glossary and return to the main translation interface. Paste your product announcement email again, select your target language, and this time make sure that the glossary option is enabled.

Notice how using the glossary ensures that “TaskFlow” never gets translated, and technical terms maintain consistency which is critical for brand recognition and professional documentation.

When to use the glossary feature in DeepL:

• Glossaries are crucial for brand language (company and product names)

• Industry-specific documentation (technical terms that need precise translations)

• Product manuals (consistent feature names across languages)

• Legal documents (specific terminology requirements)

• Marketing materials (maintaining brand voice across markets).

But what if you need to refine your English writing itself, not just translate it?

What is DeepL Write?

DeepL Write is an AI-powered writing assistant that assists in improving the tone, fluency, and precision of text. In contrast to translation tools, DeepL Write concentrates on enhancing your original English content by recommending better word choices, correcting grammatical errors, and modifying the tone overall to suit your target audience.

To use DeepL Write, navigate to the DeepL Write tab and paste the following email draft into the text box:

Hi team, I wanted to reach out about the new project timeline. We got some delays because of unexpected issues with the vendor. I think we should probably meet next week to discuss how to handle this. Let me know if you have time. Thanks. 

Copy to clipboard
Copy to clipboard

As you type or paste text, DeepL Write automatically highlights areas for improvement and suggests alternative phrasings. And you can also choose the writing style and tone from the style dropdown. Here is a sample output:

So if you’ve understood how DeepL Write works, here is a short exercise for you:

Write a professional email of your own announcing a team meeting. Run it through DeepL Write using the style and tone setting. Compare your original version with DeepL Write’s suggestions and notice improvements.

Now that you’ve seen what DeepL can do, let’s explore the technology that makes it work.

How does DeepL work?

DeepL uses neural networks based on transformer architecture, the same technology powering modern AI systems like ChatGPT. Unlike traditional translators that work word by word, DeepL interprets entire sentences contextually, understanding how words relate to each other and what they mean together.

The translation process happens in several steps:

• First, your input text is tokenized and encoded, breaking sentences into smaller units that the system can process.

• Next, the neural network generates context vectors that capture the meaning and relationships within your text, analyzing grammar, syntax, and semantic connections.

• Then, a decoder uses these context vectors to predict the natural translation in the target language, considering multiple possible phrasings before selecting the best option.

• Finally, the system learns continuously, improving its accuracy over time by training on professionally translated documents.

DeepL produces translations that read as if a native speaker wrote them originally, maintaining the tone, style, and intent of your source text.

So how does DeepL compare to the industry standard, Google Translate?

Is DeepL better than Google Translate?

When we compare DeepL and Google Translate, the answer depends on what you prioritize. Let’s examine the key parameters:

In summary, choose DeepL for quality and naturalness in supported languages, and Google Translate for broader language coverage and convenience features.

Advantages of DeepL

Here’s what makes DeepL a preferred choice for professionals and businesses:

• DeepL produces human-like translations for European and major Asian languages, capturing tone and context better than most competitors.

• It preserves structure, layouts, tables, and formatting when translating Word, PDF, and PowerPoint files.

• Helps create custom glossaries to ensure consistent terminology across all translations, essential for brand language and technical documentation.

• DeepL Write is a built-in English writing assistant refines tone, clarity, and fluency beyond just translation.

• Distraction-free, intuitive design that focuses on the translation task without unnecessary features.

• Reliable data protection with no text storage on free plans, and enterprise-grade security options for business users including SOC 2 compliance.

These advantages make DeepL a strong contender in the translation tools market, especially for quality-focused work.

Conclusion

DeepL is a powerful neural machine translation tool that delivers natural, context-aware translations for over 30 languages. By translating text and documents, creating custom glossaries for consistency, and using DeepL Write to refine English content, you can produce professional-quality translations that maintain tone, formatting, and brand voice across multiple languages, making it ideal for business communication, academic work, and international marketing.

Explore Codecademy’s Intro to Large Language Models (LLMs) to understand how neural networks and transformer architecture power tools like DeepL.

Frequently asked questions

1. What is the use of DeepL?

DeepL translates text and documents across over 30 languages while maintaining natural tone and context. It’s ideal for business communication, academic writing, and marketing materials where translation quality matters.

2. Is DeepL still free?

Yes, DeepL offers a free version with a 5,000-character limit per translation and limited document translations. Paid plans start at $8.99 per month for unlimited translations and advanced features.

3. Is ChatGPT better than DeepL?

DeepL is specifically designed for translation and produces more accurate, natural results than general-purpose tools like ChatGPT. For professional translation work, DeepL is the better choice.

4. Is DeepL considered AI?

Yes, DeepL uses neural networks and transformer architecture trained on millions of professionally translated texts to deliver contextually accurate translations.

5. Can I use DeepL on my phone?

Yes, DeepL offers mobile apps for iOS and Android, or you can access it through your mobile browser at deepl.com.