EnglishOriginal
Translation
Introducing Aure,
DeepL’s most natural voice model yet
Your rhythm, intonation and emphasis now translate with your words, in real time. Slow down to make a point. Speed up when you're excited. Lean in when it matters. Switch languages. Stay yourself.
Listen between the lines
More of your voice than ever
Adapts as you speak
Aure continuously mirrors the way you speak, reflecting shifts in tone, pace, and expression as they happen.
Sounds unmistakably you
Aure carries what makes your voice uniquely yours across languages. It’s matched live and never stored as a voice profile.
Available in 8 languages
Aure is available across DeepL Voice in English, German, Spanish, French, Italian, Japanese, Portuguese and Korean—no setup needed.
Wherever conversations happen
Translate meetings in Microsoft Teams, Zoom and Google Meet in real-time, with 76% fewer critical translation errors than platform-native translation tools.
Communicate face-to-face across languages in real time, from the front desk to the factory floor.
Embed real-time voice translation into your own apps and products with the DeepL API for Voice, making every interaction more natural, accessible and connected.
Bring your voice to every conversation.
Try DeepL’s most natural voice translation model yet.
Frequently asked questions
Aure supports eight languages: English, German, Spanish, French, Italian, Japanese, Portuguese, and Korean, with more languages coming soon. For languages not yet supported by Aure, DeepL Voice automatically uses our previous voice models.
No setup is required. Aure is automatically available for supported languages across DeepL Voice for Online Meetings, DeepL Voice for In-Person Conversations, and the DeepL Voice API.
Aure continuously analyzes how you speak in real time, using your original speech to shape how your voice sounds and expresses emotion, tone and emphasis in another language. This goes beyond traditional voice matching which typically uses a brief sample of your voice to maintain a consistent vocal identity throughout a conversation.
Yes. DeepL Voice filters out background noise, other speakers and echoes to isolate the primary speaker’s voice. This helps preserve natural-sounding translated speech even in busy environments.
No. Aure matches your voice dynamically during the live session without creating or storing a permanent voice profile. Voice data is processed temporarily in memory and deleted when the session ends.
DeepL does not permanently store transcription or translation data. Meeting data is also processed temporarily and deleted when the session ends. Data is encrypted in transit and is not used to train DeepL’s models.