基于两阶段进化搜索的精确金属板混响参数估计
Accurate Plate Reverb Parameter Estimation Using Two-Stage Evolutionary Search
浏览论文内容
中文总结 AI 辅助
针对第一届DAFx参数估计挑战赛任务A,提出两阶段进化搜索方法,用CMA-ES和三元搜索分别估计金属板混响器的六个物理参数,在50个IR验证集上取得良好效果。
中文摘要 AI 辅助
本文介绍了我们对第一届DAFx参数估计挑战赛任务A的参赛方案。该任务是从单个脉冲响应(IR)中恢复模拟金属板混响器的六个物理参数,包括其尺寸和材料属性。我们将此问题视为黑盒优化:将候选参数集输入模拟器,并通过与目标IR的损失函数进行评分。该方法分为两个阶段:第一阶段使用进化优化器CMA-ES恢复六个参数中的五个,在幅度归一化损失下比较IR;幅度归一化使搜索更鲁棒,但丢失了第六个参数(板的表面密度)的线索,因此第二阶段在未归一化损失上使用三元搜索单独估计该参数。由于损失函数的选择会强烈影响搜索,我们预先选择了损失函数,并分析了常见多尺度频谱损失中的压缩为何会降低恢复效果。最后,我们在包含50个IR的验证集上测试了我们的方法,讨论了一种病态失败模式,并进行了消融实验以证明采用两个不同阶段而非统一CMA-ES搜索的合理性。
英文摘要
We describe our submission to Task A of the 1st DAFx parameter estimation challenge. The task is to recover the six physical parameters of a simulated metal-plate reverberator -- its dimensions and material properties -- from a single impulse response (IR). We treat this as a black-box optimization: candidate parameter sets are fed to the simulator and scored by a loss against the target IR. The method has two stages. The first uses CMA-ES, an evolutionary optimizer, to recover five of the six parameters, comparing IRs under an amplitude-normalized loss. Amplitude normalization makes the search robust but discards the cue to the sixth parameter, the plate's surface density; a second stage therefore estimates it alone, with a ternary search on the un-normalized loss. As the choice of loss strongly affects the search, we select it beforehand, and analyze why compression in the common multi-scale spectral loss degrades recovery. Finally, we test our method on a validation set of 50 IRs, discuss a pathological failure mode, and ablate to justify having two different stages instead of a unified CMA-ES search.
发表机构
- Siebel School of Computing and Data Science(西贝尔计算与数据科学学院)
- University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校)
机构由 AI 辅助整理,请以论文原文为准。