arXiv：cs.AI· Chenlong Wei, Hanling Wang, Mingjin Lou, Yufan Shi, Yandong Sun, Sheldon Yu, Xiaohui Zhu, Xiaobo Jin, Yiwei Wang·· 10 小时前
SonicWorld：面向空间音频推理的以声源为中心的潜在动态模型
SonicWorld: Source-Centric Latent Dynamics for Spatial Audio Reasoning
AI 导读
SonicWorld 将物理、学习到的声学与语义特征整合为持久声源状态，并投影进 Omni LLM 进行空间推理。配套 Sonic-Bench 包含 2.8K 一阶 Ambisonics 片段与 28.8K 目标问答对，覆盖 480 个仿真室内场景。
正文
Abstract:Dynamic spatial audio reasoning requires a model to associate changing acoustic observations with the correct source and recover its spatial state at a queried time or interval, a challenge intensified by overlapping speech and source motion. To study this problem, we introduce Sonic-Bench, comprising 2.8K first-order Ambisonics clips and 28.8K target-specific question-answer pairs across 480 simulated indoor scenes. The benchmark covers localization, distance estimation, and motion reasoning in single- and dual-source settings. We further propose SonicWorld, which integrates physical, learned acoustic, and semantic features into persistent source states through learned transitions and recurrent updates, and projects them into an Omni LLM for spatial reasoning. Compared with Spatial-Omni on the Sonic-Bench test split, SonicWorld improves direction accuracy from 31.98% to 34.69% and reduces azimuth error from 57.22° to 48.01°. In zero-shot transfer to Spatial LibriSpeech, SonicWorld retains strong localization performance, achieving 73.98% direction accuracy and 19.29° azimuth error, compared with 27.14% and 70.74° for Spatial-Omni.
| Comments: | 37 pages, 12 figures, including appendices |
| Subjects: | Sound (cs.SD); Artificial Intelligence (cs.AI); Audio and Speech Processing (eess.AS) |
| Cite as: | arXiv:2610.17131 [cs.SD] |
| (or arXiv:2610.17131v2 [cs.SD] for this version) | |
| https://doi.org/10.48550/arXiv.2610.17131 arXiv-issued DOI via DataCite |
Submission history
From: Chenlong Wei [view email]
[v1]
Mon, 12 Oct 2026 17:19:19 UTC (8,412 KB)
[v2]
Wed, 14 Oct 2026 07:24:59 UTC (8,412 KB)
来源：arXiv：cs.AI · arxiv.org