Tech Meridian ← LIVE FEED
PROMY MERIDIAN RU

NEWS · MODELS · #3422

DeepL's Aure model brings tone, emotion and pace to real-time translation

DeepL announced Aure, a new model for DeepL Voice that continuously analyzes original audio and translated text to transfer tone, emotion, emphasis, and speaking pace into the target language in real time. Aure runs automatically in DeepL Voice for Online Meetings, In-Person Conversations and via the DeepL API, and initially supports English, German, Spanish, French, Italian, Japanese, Portuguese and Korean.

KEY POINTS

  1. DeepL announced Aure, a new model for DeepL Voice that continuously analyzes original audio and translated text to transfer tone, emotion, emphasis, and speaking pace into the target language in real time.
  2. Aure runs automatically in DeepL Voice for Online Meetings, In-Person Conversations and via the DeepL API, and initially supports English, German, Spanish, French, Italian, Japanese, Portuguese and Korean.
  3. Making translated speech preserve mood, emphasis and pace can materially improve intelligibility and conversational naturalness in multilingual real-time communication.

WHY IT MATTERS

Making translated speech preserve mood, emphasis and pace can materially improve intelligibility and conversational naturalness in multilingual real-time communication.

SOURCES & TIMELINE

1