arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2602.17732eess.AScs.SDeess.SP
收藏

SIRUP: 一种基于扩散的虚拟上混器,用于高定向空间化的第一阶ambisonics

SIRUP: A diffusion-based virtual upmixer of steering vectors for highly-directive spatialization with first-order ambisonics

1Center for Advanced Intelligence Project, RIKEN, Japan 2Sorbonne University, France 3LTCI, T\' e l\' e com Paris, Institut Polytechnique de Paris, France 4Graduate School of Engineering, Kyoto University, Japan -1mm

Emilio Picard, Diego Di Carlo, Aditya Arie Nugraha, Mathieu Fontaine, Kazuyoshi Yoshii

AI总结:

SIRUP通过潜在扩散模型实现基于第一阶ambisonics的虚拟上混器,提升高定向空间化性能。

摘要:

本文提出了一种通过更少通道的球形麦克风阵列捕获的虚拟上混器。该挑战传统上通过从第一阶ambisonics(FOA)数据中恢复声源的方向和信号,然后使用基于物理的声学模拟器生成更高阶ambisonics(HOA)数据来解决。然而,这种方法在处理声源估计的空间定向性与FOA ambisonics数据的空间分辨率之间的相互依赖性时存在困难。我们的方法,称为SIRUP,采用了一个潜在扩散模型架构。具体来说,使用变分自编码器(VAE)学习HOA数据在潜在空间中的紧凑编码,然后训练扩散模型生成HOA嵌入,这些嵌入由FOA数据条件化。实验结果表明,SIRUP在针对指向向量上混器、源定位和语音去噪方面相比FOA系统实现了显著的改进。

英文摘要:

This paper presents virtual upmixing of steering vectors captured by a fewer-channel spherical microphone array. This challenge has conventionally been addressed by recovering the directions and signals of sound sources from first-order ambisonics (FOA) data, and then rendering the higher-order ambisonics (HOA) data using a physics-based acoustic simulator. This approach, however, struggles to handle the mutual dependency between the spatial directivity of source estimation and the spatial resolution of FOA ambisonics data. Our method, named SIRUP, employs a latent diffusion model architecture. Specifically, a variational autoencoder (VAE) is used to learn a compact encoding of the HOA data in a latent space and a diffusion model is then trained to generate the HOA embeddings, conditioned by the FOA data. Experimental results showed that SIRUP achieved a significant improvement compared to FOA systems for steering vector upmixing, source localization, and speech denoising.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑