Problem
Research questions and friction points this paper is trying to address.
out-of-domain
loss functions
parameter loss
partial parameter distance
sound-matching
Evaluating Loss Functions in Differentiable Out-of-Domain Sound-Matching with Partial Parameter Distance
🤖 AI Summary
研究通过引入部分参数距离(PPD)方法，解决了不同领域声音匹配中损失函数评估的问题，并在多种合成场景下验证了其有效性。
📝 Abstract
In out-of-domain (OOD) sound-matching, a synthesizer is optimized to mimic a sound it did not generate. OOD evaluation of loss functions is underexplored in part because the standard "parameter loss" metric requires a shared parameter space between target and imitator, which OOD settings lack. We introduce Partial Parameter Distance (PPD), which applies parameter loss only to the critical parameters that mismatched synthesizers share (e.g., filter cutoffs), enabling automatically evaluated OOD experiments; we verify its results with blinded listening tests. Across seven scenarios involving band-pass filtering, amplitude modulation, and pitch-bending, we evaluate four differentiable loss functions (SIMSE_Spec, L1_Spec, JTFS, DTW_Envelope). Loss-function effectiveness remains tightly coupled to the method of synthesis: SIMSE_Spec excels at filter-cutoff recovery, DTW_Envelope at amplitude-modulation recovery, and JTFS at smooth pitch trajectories. Parameter-based evaluation agrees with listening tests on the top-ranked loss function in five of seven scenarios, demonstrating its utility as a diagnostic tool.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Partial Parameter Distance
Out-of-Domain Sound-Matching
Loss Function Evaluation
🔎 Similar Papers
No similar papers found.
