DeepL released its voice model, Aure.
The story so far
DeepL has released the voice model Aure, which conveys the speaker's tone, emotion, and pace to the target language by analyzing the original audio and translated text. Aure supports eight languages, including English, German, and Spanish, and is suitable for DeepL Voice online meetings, face-to-face conversations, and the DeepL API. The uniqueness of Aure lies in its continuous analysis of the translated text and original audio during the conversation, and the separation of voice identity, speech melody, and content during training, allowing each component to be independently translated into the new language. Additionally, reinforcement learning is used to eliminate accents in the output.
AI-written from the reports · Updated 3 days ago
Coverage timeline
Follow the coverage to see every side of the story.
- ITHome · AIDeepL launches its voice AI translation model Aure, reproducing speaker's tone and emotion in real-time
DeepL has launched the voice AI translation model Aure, which reproduces the speaker's tone and emotions in real-time. Aure is based on DeepL's voice preservation technology and can dynamically adapt the translated voice, capturing the fluctuations in tone, emotional shifts, emphasis, and changes in speech rate throughout the process, supporting eight languages including English and German.
- The DecoderThe DeepL Aure model achieves real-time translation that conveys tone and emotion.
The DeepL Aure model conveys the speaker's tone, emotion, and pace to the target language by analyzing the original audio and translated text, supporting eight languages including English, German, and Spanish, and is suitable for DeepL Voice online meetings, face-to-face conversations, and the DeepL API.
Heat over time
- Now
- 3
- Peak
- 1911 Nov, 23:00
- 24h change
- -75%
Hover or tap the chart to see each hour; use the arrow keys on a keyboard.