Corpus ID: 219792428

Boosting Objective Scores of Speech Enhancement Model through MetricGAN Post-Processing

@article{Fu2020BoostingOS,
  title={Boosting Objective Scores of Speech Enhancement Model through MetricGAN Post-Processing},
  author={S. Fu and Chien-Feng Liao and Tsun-An Hsieh and Kuo-Hsuan Hung and S. Wang and Cheng Yu and Heng-Cheng Kuo and Ryandhimas E. Zezario and Y. Li and Shang-Yi Chuang and Y. Lu and Y. Tsao},
  journal={ArXiv},
  year={2020},
  volume={abs/2006.10296}
}

S. Fu, Chien-Feng Liao, +9 authors Y. Tsao

Published 2020

Computer Science, Engineering

ArXiv

The Transformer architecture has shown its superior ability than recurrent neural networks on many different natural language processing applications. Therefore, this study applies a modified Transformer on the speech enhancement task. Specifically, the positional encoding may not be necessary and hence is replaced by convolutional layers. To further improve PESQ scores of enhanced speech, the L_1 pre-trained Transformer is fine-tuned by MetricGAN framework. The proposed MetricGAN can be… CONTINUE READING

View PDF on arXiv

Save to Library

Create Alert

Cite

Launch Research Feed