DeepL Voice Receives High Marks for Translation Quality and Subtitle Stability in Independent Benchmark Study by Slator

Cologne, Germany – March 25, 2026 – DeepL, a global company specializing in AI products and research, today announced the results of an independent benchmark study conducted by Slator, a firm specializing in research and market analysis for the translation, localization, interpretation, and language AI industries. This study compared the quality and stability of real-time AI translation and captioning features on major online meeting platforms such as Google Meet, Microsoft Teams, and Zoom.

As multilingual meetings with global teams become commonplace, companies are increasingly relying on AI voice translation technology to support collaboration, customer negotiations, and cross-border decision-making. This study, commissioned by DeepL and independently evaluated and analyzed by Slator, found that DeepL Voice achieved the highest scores for both translation quality and subtitle stability, outperforming the native captioning features in Google Meet, Microsoft Teams, and Zoom.

In high-stakes situations like sales negotiations, supplier discussions, and strategic management meetings, even minor mistranslations or subtitle instability can disrupt the flow of discussion, cause confusion, and degrade the quality of decision-making. As global business accelerates, AI translation is no longer a 'nice-to-have' feature but is becoming a foundational infrastructure for operations. The strength of that foundation depends on two factors: translation quality and the stability of live captions.

Key findings from the study include:

• Highest score for translation quality: In human evaluations, DeepL Voice (for Zoom) scored 96.4 out of 100, and DeepL Voice (for Microsoft Teams) scored 96.3. This was significantly higher than the 87-89 points scored by other evaluated platforms.

• Reduced high-severity translation errors by an average of 76%: The occurrence of 'critical' or 'major' translation errors, such as misinterpretations that could significantly affect meeting progress or decision-making, was reduced by an average of 76% compared to other platforms.

• 79% of translation segments were entirely error-free: The proportion of translated captions rated as 'fully passing' reached 79%, far exceeding the competitor average of 42%.

• High marks for subtitle stability as well: In Slator's automated frame-by-frame analysis, DeepL Voice (for Zoom) recorded a stability score of 88.6/100, and DeepL Voice (for Microsoft Teams) scored 85.8/100.

• Significantly suppressed subtitle rewriting (flicker): The phenomenon of captions frequently rewriting on the screen (caption churn) was reduced by an average of 37.6% compared to Microsoft Teams and 54.7% compared to Zoom.

• 96% of professional linguists preferred DeepL Voice in a blind evaluation: Across all blind evaluations, 96% of the professional translators and interpreters who participated chose DeepL Voice as the superior solution.

Jarek Kutylowski, CEO and founder of DeepL...