SCP-GAN: Self-Correcting Discriminator Optimization for Training Consistency Preserving Metric GAN on Speech Enhancement Tasks

26 Oct 2022  ·  Vasily Zadorozhnyy, Qiang Ye, Kazuhito Koishida ·

In recent years, Generative Adversarial Networks (GANs) have produced significantly improved results in speech enhancement (SE) tasks. They are difficult to train, however. In this work, we introduce several improvements to the GAN training schemes, which can be applied to most GAN-based SE models. We propose using consistency loss functions, which target the inconsistency in time and time-frequency domains caused by Fourier and Inverse Fourier Transforms. We also present self-correcting optimization for training a GAN discriminator on SE tasks, which helps avoid "harmful" training directions for parts of the discriminator loss function. We have tested our proposed methods on several state-of-the-art GAN-based SE models and obtained consistent improvements, including new state-of-the-art results for the Voice Bank+DEMAND dataset.

PDF Abstract
No code implementations yet. Submit your code now

Datasets


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Speech Enhancement VoiceBank + DEMAND SCP-CMGAN PESQ 3.52 # 1
CSIG 4.75 # 1
CBAK 3.97 # 1
COVL 4.25 # 1
STOI 96 # 1
SSNR 10.82 # 2

Methods


No methods listed for this paper. Add relevant methods here