HelixWorld creates real-time worlds with matching sounds and visuals
HelixWorld: A Real-time Interactive Audio-Visual World Model
Computer Vision and Pattern Recognition
Summary
Worlds in video games and simulations usually show pictures and movement but don't add matching realistic sounds that change as you move around. The authors build HelixWorld, a system where pictures and stereo sounds change together in real time based on where the user looks and moves. They collected special data linking real camera positions with sounds to train their system. Their model runs fast on one computer chip and keeps the audio and visuals perfectly synced as you interact, making the experience more immersive than before.
What this means in practice
- •For virtual reality developers: Create VR environments where the sound perfectly matches user movement and visual perspective for more immersive experiences.$Commercial implications: Enables commercial VR products with synchronized spatial audio-visual interaction that improves realism and user immersion.
- •For game engine engineers: Integrate interactive audio-visual models that maintain spatial sound consistency during camera motion in real time.
Authors
Lei Ke, Jiahao Pan, Zeyue Tian, Jiaming Wang, Haoyuan Huang, Kam Man Wu, Pengjun Fang, Hongyu Liu, Chenyang Qi, Lin Wang, Ruibin Yuan, Weijia Chen, Fangneng Zhan, Qifeng Chen, Wei Xue, Yike Guo
Abstract
World simulation is inherently multisensory, demanding synchronized visual and acoustic dynamics in real time. Yet prevailing interactive world models remain strictly silent, focusing exclusively on visual rendering and control while overlooking the acoustic dimension. We present HelixWorld, a real-time interactive audio-visual world model where visual scenes and camera-grounded spatial stereo sound co-evolve natively under user interaction. We curate a high-fidelity spatial audio-visual dataset with true stereo acoustics and metric camera poses, upon which we pre-train a bidirectional teacher conditioned on 6-DoF camera trajectories and user actions. To enable low-latency causal interaction, we distill the teacher into a few-step streaming student via an online trajectory distillation loss, sustaining drift-free joint audio-visual rollouts at 24 FPS on a single GPU. Furthermore, we formalize spatial-acoustic consistency and introduce HelixBench to evaluate whether synthesized sound fields faithfully track dynamic viewpoint motion. Extensive experiments demonstrate that HelixWorld matches state-of-the-art silent world models in visual fidelity and responsiveness, while significantly surpassing existing baselines in camera-aligned spatial-acoustic immersion.