DeepL's new Aure model brings tone, emotion and pace to real-time voice translation
The Decoder · rss · 2026-11-11
DeepL's real-time voice translation is getting more natural: the new Aure model carries over tone, emotion, emphasis, and speaking pace from the original speaker's voice into the target language, aiming to finally capture sarcasm and mood shifts in DeepL Voice.
More from Multimodal
- UniPath (NeurIPS 2026): adaptively coordinating understanding and generation in one model — jindong_wang92 · 2026-11-11
- An AI-built RPG where a PhD student battles peer reviewers — s_scardapane · 2026-11-11
- Manycore hosts Aholo Global 3D AI Hackathon with its Lux3D 3D-generation model — Scobleizer · 2026-11-11
- Same Seedance 2.5 model costs $0.473/s on fal but $0.208/s on YouArt Desktop — FellMentKE · 2026-11-11
- Two days after launch, Fable 5.2 users are churning out AI-made videos — iamfakhrealam · 2026-11-11
- Meng To turns his three.js game into a 60-second trailer with Fable 5.2 — iamfakhrealam · 2026-11-11