Shizuku-Duplex Launches as First Full-Duplex Audio Synthesis Model for AI Characters
venturetwins · x · 2026-11-08
ShizukuAILabs introduced Shizuku-Duplex, billed as the world's first full-duplex audio synthesis model built specifically for AI characters, enabling natural listening-while-speaking interaction. The team also shared a video explaining how their models are designed for natural human-AI interaction.
More from Multimodal
- Full GPT Image 2 prompt template for pastel paper-collage city posters — SimplyAnnisa · 2026-11-08
- Open-source vocal synth models vocal tract as acoustic tube network, adds MusicXML support — TinfoilTricorn · 2026-11-08
- Cat Steals Package, Two Drones Give Chase in Seedance 2.5 AI Short — nikola_mr64990 · 2026-11-08
- Building a deterministic ComfyUI game-art pipeline with an orchestrator agent: which steps should be scripted? — zimmer550king · 2026-11-08
- One character across 8 isometric angles: ControlNet, spritesheet LoRA, or image-to-3D rerender? — zimmer550king · 2026-11-08
- Open-source Prompt Vault v7.1 stores prompts locally with ComfyUI node metadata auto-import — danger_uxb · 2026-11-08