GDC Festival of Gaming is part of the Informa Festivals Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

March 1-4, 2027
Moscone CenterSan Francisco, CA

Agenda

Neural Spatial Audio in 'Roco Kingdom'

Maode Shi  (Game Engine Specialist, Tencent Games)
YITING LI  (Technical Audio Designer, Tencent Games)
Pass Type: Academic Pass, Festival Pass, Game Changer Pass, Indie Pass, Start-Up Pass - Get your pass now!
Track: Programming, Audio
Audience Level: Intermediate
Format: Lecture
Vault Recording: TBD
Audience Level: Intermediate
Large game worlds need convincing acoustics, but detailed sound propagation can be expensive to store and render, while manually authoring rooms and portals can add substantial production effort. This session presents a new neural acoustics system developed for Roco Kingdom, combining improved acoustic fidelity, directional and multiband support, reduced manual authoring, and practical performance on PC and mobile. The session shares the design reasoning, experiments, and production experience behind the system, including the trade-offs involved in preserving acoustic detail within limited memory and runtime budgets. These lessons extend beyond acoustics to developers using neural networks to encode data in other game systems.

Takeaway

Attendees learn how a new neural acoustics system delivers directional, multiband sound on PC and mobile. They gain insight into the design decisions and trade-offs that make neural data encoding practical in games, with lessons applicable to other neural systems constrained by quality, memory, and runtime budgets.

Intended Audience

This session is intended for audio programmers and audio technical artists interested in neural acoustics and its application in game development. Attendees are expected to have a basic understanding of acoustics. Familiarity with neural network fundamentals is helpful but not required.