Back to events
Design
Score 70Heat 0
DeepL introduces Aure model to translate intonation and speaking speed in real time
LaunchSpeechMultimodalDeepLMixhalo
Summary
DeepL has launched a new real-time translation voice model called Aure that preserves tone, emotion, emphasis, and speaking pace. The model continuously analyzes original audio and translated text to capture mood shifts and sarcasm.
Why it matters
Advances real-time speech translation capabilities by incorporating prosody and emotional context into cross-lingual communication.
Editorial score
2 reviews6.5
Significance
7.5
Novelty
8.5
Credibility
6.5
Reach
6
Actionability