Back to news
launchDeepL2026-11-11

DeepL launches Aure voice translation model reproducing tone and emotion shifts in real time, without Chinese support

DeepL launched the Aure voice AI translation model on November 11, the engine behind DeepL Voice, reproducing speaker tone, emotion and pace in real time, initially supporting eight languages with no Chinese.

DeepL launched the Aure voice AI translation model on November 11 local time, which the company calls its most dynamic and realistic voice model to date, though it does not support Chinese for now.

As the engine behind DeepL Voice, Aure continuously adapts the translated speech in real time, reproducing the speaker's tonal rises and falls, emotional shifts, stress emphasis and pace, so voice translation conveys not just literal words but the individuality, subtle emotion and expressive richness that give conversation its naturalness. The company said Aure builds on DeepL's recently introduced voice preservation technology, which lets people communicate across languages in their own highly recognizable original timbre. Aure goes further by capturing the dynamic movement and immediate reactions of a voice throughout a conversation, injecting more spontaneity and authenticity from natural human speech into real-time translation.

Aure is already supported in DeepL Voice for web conferencing, DeepL Voice for face-to-face conversations, and the DeepL Voice API. The model launched with support for English, German, Spanish, French, Italian, Japanese, Portuguese and Korean, and will be extended to more languages over time, with DeepL Voice automatically falling back to its previous voice model for languages not yet supported.

DeepLAure语音翻译实时翻译DeepL Voice