How can you know when to trust an AI translation? It’s a question that DeepL and our customers spend a lot of time discussing. We've built a solution that earns that trust, and gives you the confidence you need.
Our new Translation Quality Evaluation (TQE) AI model detects potential errors in DeepL translations and flags them to human reviewers. Because it’s a different model to the one that generates translations, it’s not a case of AI “marking its own homework.” It’s a feature that works with human reviewers so they can confirm the quality of their translations in a fraction of the time it would usually take.
Search DeepL AI Labs and you’ll find our full TQE demo and details for how you can start experimenting in our early access playground.
00:00
