.png){kind=icon}
返回
Gradient-Aware Loss Function for Improved Learning in Speech Enhancement
.png){kind=icon}
PRE
AI
DOI：10.1109/LSP.2026.3657709.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Conventional loss functions for gradient descent are designed mainly to assess output quality, with limited attention to gradient behavior. This study identifies the gradient inconsistency problem in existing loss functions, where gradient magnitude is misaligned with the loss value, causing unstable learning and poor correspondence between optimization and the target metric. To resolve this, we propose a gradient-aware loss function whose gradient is proportional to a chosen evaluation metric, with the loss derived through integration. This design ensures consistent gradients, stable convergence, and alignment with the evaluation metric. Applied to speech enhancement using the log-spectral distance metric, the method achieves faster convergence, greater stability, and higher perceptual quality than conventional approaches, validating the superiority of the proposed formulation.
Keyword:
Gradient descent
loss function
machine learning
speech enhancement
期刊
I
IF:
3.9
论文数:
824
被引数:
0
机构
引用论文
Perceptually-motivated Environment-specific Speech Enhancement.png)
ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)IF0
PREAI
分享收藏分享收藏
查看更多内容