deep learning approaches for black-box modelling of audio effects have shown
promise, however, the majority of existing work focuses on nonlinear effects
with behaviour on relatively short time-scales, such as gu
通过比较 State Space 模型、Linear Recurrent Units 和 Long Short Term Memory 网络在模拟音频效果方面的性能,本文研究了近期机器学习进展在虚拟类比建模中的应用,包括信号历史编码、能量包络、频率内容和瞬态等方面的准确度。结果表明,Long Short Term Memory 网络在模拟失真和均衡器方面的准确度较高,而 State Space 模型在饱和和压缩方面的模拟能力超过其他方法。对于长时间变化特性,State Space 模型展现了最高的准确度。Long Short Term Memory 网络和 Linear Recurrent Unit 网络则更容易引入音频伪像。