Problem
Research questions and friction points this paper is trying to address.
Game sound-effect completion
Audio inpainting
Acoustic identity
Event-level transformation
Game Sound-Effect Completion with Event-Level Transformation Hints
🤖 AI Summary
This study addresses the challenge of game skin sound effect inpainting, which requires balancing the preservation of event-character consistency with differentiated reconstruction. To this end, we propose a latent diffusion inpainting model fine-tuned on Stable Audio, incorporating a symbolic soft retention mask and an adjustable event-level transformation prompting mechanism. Coupled with a League of Legends audio data alignment pipeline, this approach enables flexible control over the trade-off between original feature retention and redesign proportion. Experimental results on a held-out test set demonstrate that our method significantly outperforms general-purpose audio editors in reconstruction quality. Furthermore, target-derivative prompts effectively enhance pairwise similarity, achieving high-quality and controllable game sound effect inpainting.
📝 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.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Sound-Effect Completion
Latent Inpainting
Transformation Hints
Signed Soft Retention Mask
Conditional Generation
🔎 Similar Papers
No similar papers found.
