English

MetricGAN+: An Improved Version of MetricGAN for Speech Enhancement

Sound 2021-06-07 v2 Artificial Intelligence Audio and Speech Processing

Abstract

The discrepancy between the cost function used for training a speech enhancement model and human auditory perception usually makes the quality of enhanced speech unsatisfactory. Objective evaluation metrics which consider human perception can hence serve as a bridge to reduce the gap. Our previously proposed MetricGAN was designed to optimize objective metrics by connecting the metric with a discriminator. Because only the scores of the target evaluation functions are needed during training, the metrics can even be non-differentiable. In this study, we propose a MetricGAN+ in which three training techniques incorporating domain-knowledge of speech processing are proposed. With these techniques, experimental results on the VoiceBank-DEMAND dataset show that MetricGAN+ can increase PESQ score by 0.3 compared to the previous MetricGAN and achieve state-of-the-art results (PESQ score = 3.15).

Keywords

Cite

@article{arxiv.2104.03538,
  title  = {MetricGAN+: An Improved Version of MetricGAN for Speech Enhancement},
  author = {Szu-Wei Fu and Cheng Yu and Tsun-An Hsieh and Peter Plantinga and Mirco Ravanelli and Xugang Lu and Yu Tsao},
  journal= {arXiv preprint arXiv:2104.03538},
  year   = {2021}
}

Comments

Accepted by Interspeech 2021