Game sound effects completed with event-level redesign guidance
Game Sound-Effect Completion with Event-Level Transformation Hints
Sound
Summary
When making new sound effects for game characters, it’s important that the sounds match the game’s actions but still sound unique. The authors created a method that fills in missing sounds for a character’s skin by using existing sounds, completed sounds, and design instructions. Their model can choose how much to keep from the original sound versus redesign it, making the new sounds fit well with gameplay. Tests showed their method works better than general sound editing tools.
What this means in practice
- •For game audio designers: Create consistent sound effect sets for new game character skins with customizable redesign levels for each event.$Commercial implications: Enables a product that automates tailored audio creation for game skins, saving time and improving design consistency.
- •For interactive media developers: Automatically generate missing audio events in interactive applications by conditioning on partial audio data and event descriptions.
Authors
Xinrui Jiang, Heng Yu
Abstract
Creating sound effects for a new game-character skin requires a distinct acoustic identity while preserving gameplay-event roles. The challenge is to complete a coherent set of related sounds whose required degrees of redesign differ. We formulate this task as completion conditioned on base-skin audio, completed target assets, and a textual design description. We develop a pipeline to collect, process, and align corresponding events across League of Legends skins. Building on Stable Audio 3's pretrained audio prior, we fine-tune a latent inpainting model to jointly complete missing events. A signed soft retention mask encodes available audio and an adjustable transformation hint for each missing event, specifying the requested balance between retention and redesign. Experiments on held-out skins show improved reconstruction over the evaluated general-purpose audio editors. Target-derived hints further improve paired similarity, with three-level hints retaining most of the benefit of continuous guidance.